PATH-DEPENDENT DEADLOCK: THE INSTITUTIONAL CAUSES OF THE EURO CRISIS

From the SelectedWorks of Samuel Dahan Dr. August 18, 2015 PATH-DEPENDENT DEADLOCK: THE INSTITUTIONAL CAUSES OF THE EURO CRISIS Samuel Dahan, Dr. A...
0 downloads 3 Views 2MB Size
From the SelectedWorks of Samuel Dahan Dr.

August 18, 2015

PATH-DEPENDENT DEADLOCK: THE INSTITUTIONAL CAUSES OF THE EURO CRISIS Samuel Dahan, Dr.

Available at: http://works.bepress.com/samuel_dahan/1/

1

A PATH-DEPENDENT DEADLOCK: INSTITUTIONAL CAUSES OF THE EURO CRISIS

Samuel Dahan* Abstract We argue that the characterization of the financial turmoil in the European Monetary Union as merely a sovereign debt crisis is inaccurate insofar as the deterioration of public finances represents the culmination of a process: legal and institutional flaws laid the ground for a public debt crisis. The institutional failure was the embedding of a unified monetary policy accompanied by a failure to introduce a coordinated fiscal and labor policy. The latent consequences of these flaws were brought into full view with the onset of the Global Financial Crisis. There was a lack of real convergence between the countries of the core (Germany) and the periphery (especially Greece, Ireland, Italy, Portugal and Spain, the so-called GIIPS.) This was exacerbated by the “one-size-fits-all” monetary policy, distorted fiscal policy, and the lack of a common approach to wage determination.1 In sum, instead of fostering learning and convergence, the asymmetric EMU design catapulted the GIIPS and Germany onto divergent growth paths, which ultimately translated into “destabilizing macroeconomic imbalances.” This analysis will draw upon knowledge-production systems and learning theories in order to contribute to a deeper understanding of the EMU crisis. In particular, we take a contextualized approach by examining the pathdependent evolution of the EMU and how institutional failures have turned the GFC into a sovereign debt crisis.

INTRODUCTION Identifying the cause of a problem is critical to solving it. An understanding of the policy failures that have given rise to the current financial crisis in Europe is a prerequisite for developing suitable remedies. However, the ongoing economic turmoil in the European and Monetary Union (EMU), and more specifically in Greece, suggests that a lack of consensus regarding the root cause of the crisis is hampering the emergence of an adequate solution for the EMU. It is generally agreed that the euro crisis was triggered by the Global Financial Crisis (GFC) of 2007, which began with the failure of two major American financial institutions that had speculated in unsafe mortgage-backed securities. In September 2008, Congress passed an emergency plan to bail out the financial sector, touching off *

Adjunct Professor of Law at Cornell University and Emile Noël Fellow at NYU School of Law. The author would like to thank Lynn Stout, Simon Deakin, Saule Omarova, Robert Hockett, Mark Freedland and Niklas Bruun for their support and advice during the preparation of this article, portions of which were drawn from the author’s doctoral dissertation. 1 S. Deakin, F. Wilkinson & C. Didry, Marchés du Travail, Crise Financière et Réforme: Projet d’Agenda pour une Politique du Travail, 182 HOMME SOCIÉTÉ 25–52 (2012); A. CALLINICOS, BONFIRE OF ILLUSIONS: THE TWIN CRISES OF THE LIBERAL WORLD (2013).

2 a global financial crisis that set the stage for widespread bank failure, including that of European banks. While the immediate trigger of the GFC is readily apparent, it is more difficult to understand why the international bond market began to price the growing risks associated with the debt of the EU periphery in 2009, that is to say, how the financial crisis turned into a debt crisis. We argue that the EMU crisis is more than an economic crisis; it is also a legal and institutional crisis of the EMU architecture itself. This article takes a pathdependent approach to draw a distinction between the causes of the GFC and the causes of the euro crisis, the latter being institutional. According to one explanation of the GFC, it is the result of the exponential financialization2 of the economy and growing inequality, a process strongly encouraged by governments through inadequate (de)regulation of financial services; international mobility of financial capital; and labor market flexibilization.3 While deregulation and financialization were also problems in Europe, they affected only the deficit countries, namely Spain, Ireland, Greece, Italy, and, to a lesser extent, Portugal; they did not pose problems in surplus countries such as France, Germany, Belgium and the Netherlands. Indeed, this analysis will demonstrate that while the GFC and the eurozone crisis have similar roots, the deeper cause of the eurozone crisis was the asymmetrical design of the EMU, which exacerbated the effects of the GFC in the eurozone. It is the asymmetric and procyclical macroeconomic framework established in Maastricht that we regard as the main cause of the euro crisis.4 The institutional failure was the embedding of a unified monetary policy without simultaneously introducing a coordinated fiscal and labor policy. In particular, the EMU legal framework did away with Member States’ critical instruments of macroeconomic management. Monetary policy is an exclusive Union competence vested in the ECB (European Central Bank), which set a single interest rate in order to ensure price stability in the form of an inflation target (around 2%). Fiscal policy is a shared competence with the Stability and Growth Pact (SGP), which shapes it by referring to the formal convergence criteria, with Member States retaining control over the means to the ends of the criteria. As for social policy, it remained mostly un-harmonized. Even in areas that had implications for economic policy – wage determination, for instance – there was no EU competence for wage-bargaining coordination in the The term “financialization” may be defined as growth in the scope and importance of the role of finance in capitalist economies; however, “financialization” has several meanings.2 The first is the reliance of the industry on bank loans as a source of finance. The second defines finance as an increasingly autonomous field, while the third interpretation defines it as the degree to which the network of finance extends to a wider range of actors that were originally not deeply involved in the financial sector. These dimensions of financialization are not mutually exclusive, but should be interpreted through the prism of “profit financialization,” which recognizes the maximization of shareholder value as the driving force of corporate-listed companies. For further details see Deakin, Wilkinson, and Didry, supra note 2. 3 G. Fontana, A. Rodriguez Gil & M. Sawyer, The Macroeconomic Analysis of Financialisation and Wage Inequalities, (University of Leeds Unpublished Paper, Nov. 2012) available at http://fsaraceno.free.fr/INEQ/fontana_rodriguezgil_sawyer.pdf (last visited Aug 12, 2014); D. A. Zalewski & C. J. Whalen, Financialization and Income Inequality: A Post Keynesian Institutionalist Analysis, 44 J. ECON. ISSUES 757–777 (2010). 4 C. Barnard, The Financial Crisis and the Euro Plus Pact: A Labour Lawyer’s Perspective, 41 IND. LAW J. 98–114 (2012); T. Palley, Europe’s Crisis without End: The Consequences of Neoliberalism, 32 CONTRIB. POLIT. ECON. 29–50 (2013); S. Deakin, Social Policy, Wage Determination and EMU: Towards an Egalitarian Solution to the Crisis, (2013); R. Janssen, European economic governance: The next Big Hold-up on Wages, in THERE IS AN ALTERNATIVE: ECONOMIC POLICIES AND LABOUR STRATEGIES BEYOND THE MAINSTREAM 41–43 (N. Pons-Vignon ed., 2011). 2

3 Treaty of Maastricht or Amsterdam. The latent consequences of these flaws were brought into full view with the onset of the GFC. The eurozone was vulnerable mainly for two reasons: Firstly, the EU had no mechanism for dealing with the crisis more effectively. Secondly, there was a lack of real convergence or learning between the countries of the core and the countries of the periphery, which is to say they took diverging growth paths. This was exacerbated by the “one-size-fits-all” ECB interest-rate policy, distorted fiscal learning, and the lack of either an integrated social policy or a common approach to wage determination.5 In sum, instead of fostering learning and convergence, the asymmetric design of the EMU has exacerbated divergence in growth models, which ultimately translated into “destabilizing macroeconomic imbalances.”6 On the one hand, the periphery has become highly dependent on an expansion of private credit and on increasing asset prices in the market for commercial and residential investments. On the other hand, the core has followed policies of public support for training; labor force upgrading and wage moderation; and export-led growth dependent on targeted investment in capital goods. In order to convey the complexity of the EMU flaws, this article will first reconstruct the legal and normative concepts upon which the EMU has been shaped. Our theoretical reconstruction is based on the idea that although the integration project is a multi-layered process that combines monetary, fiscal and social aspects, the social dimension has been relegated to a secondary priority. 7 While the integration project was originally conceived as an open project capable of achieving noneconomic objectives such as social welfare, the actual extent of its commitment to social aims has always been in question, and it has proven difficult to assess the treaties’ intended balance between purely economic objectives and social goals. This analysis will draw upon knowledge-production systems and learning theories, as these can help to cultivate a deeper understanding of the EMU crisis. In particular, we take a contextualized approach by looking at the path-dependent evolution of the EMU and how institutional failures have turned the GFC into a sovereign debt crisis. The remainder of this article is organized as follows: the first section offers an overview of the theoretical institutional model employed in this project (I). The second section investigates the multilayered and path-dependent institutional structures on which the EMU asymmetric framework was built.8 While social integration has always been a strong component of the EU integration project, it remains a secondary concern, subordinated to fiscal and monetary priorities (II). The third section considers the extent to which path dependency, and more precisely the asymmetric legal framework, contributed to the development of macroeconomic imbalances and, ultimately the euro crisis (III). Finally, the conclusion offers recommendations and briefly9 explores new avenues for further research, namely 5

Deakin, Wilkinson, and Didry, supra note 2; CALLINICOS, supra note 2. European Commission, Macroeconomic Imbalance Procedure - European Economic Governance, May. 17, 2013. 7 S. Deakin, The Lisbon Treaty, the Viking and Laval Judgments and the Financial Crisis, in THE LISBON TREATY AND SOCIAL EUROPE 19–45 (N. Bruun, K. Lörcher, & I. Schömann eds., 2012). 8 F. W. Scharpf, The European Social Model, 40 JCMS J. COMMON MARK. STUD. 645–670 (2002). 9 Professor J. Habermas, Democracy, Solidarity and the European Crisis, Address at KU Leuven (May. 3, 2013). 6

4 whether the EMU allows some legal flexibility to trigger the institutional shift that would break the path-dependent EMU flaws (IV). I. THE INSTITUTIONAL LEARNING FRAMEWORK Since its creation, the EMU has demonstrated slow institutional progress, and mainstream economic theories cannot fully account for the particular form of institutional change that has taken place. This research therefore relies heavily on institutional learning theories insofar as they may explain why institutional progress has been so limited since Maastricht, and how the EMU’s institutional limitations have laid the groundwork for the crisis. In particular, the institutionalist approach can be used to understand the role of pre-existing national limitations in the creation of a flawed currency union. Furthermore, it may be useful to investigate whether a substantive analysis of the path-dependent legal and institutional flaws of the EMU could trigger a more genuine learning process. A. The Institutionalist Approach The institutionalist approach is used to describe the institutional context and the evolution of institutions. The core assumption is that “different degrees of policy learning are the result of incentive and preference structures that, from a national point of view, make the transfer of policies rational.”10 Accordingly, we argue that any learning assessment should examine pre-existing historical institutional structures. According to the “institutionalist approach,” policy change is, after all, a highly political process, one that is filtered through (and can be hampered by) both path dependencies and actors’ preferences. Path dependency means that those institutions that guide decision-making reflect historical experience. As a result, preexisting institutional structures determine the limits of possible change, and decisionmakers tend to make decisions that lead only to marginal changes to the status quo. Within these parameters, the possibilities of institutional change within the EMU regime are rather limited. Change is reduced to a perpetuation of past trajectories.11 An institutionalist approach therefore conceives of the EMU regime as a closed area impervious to change. In sum, learning comes up against differences in rules, procedures and norms as well as cultural and cognitive understandings. However, institutional barriers may be overcome under conditions that create incentives for learning. B. Incentives for Institutional Learning There are various contextual conditions under which it is rational for institutions to evolve. The first condition is policy failure, namely circumstances in which policymakers realize that a policy is failing. Failure may be caused by external factors such as a crisis, which functions as a catalyst for non-routine policy learning and tends to 10

M. Citi & M. Rhodes, New Modes of Governance in the European Union: A Critical Survey and Analysis, in HANDBOOK OF EUROPEAN UNION POLITICS 463–482 (K. Erik Jørgensen, M.A. Pollack, & B. Rosamond eds., 2006). 11 C. Crouch & H. Farrell, Breaking the Path of Institutional Development? Alternatives to the New Determinism, 16 RATION. SOC. 5–43 (2004).

5 permit the mobilization of extraordinary resources.12 Indeed, research has suggested that people learn more from failure than from success.13 Studies of organizational behavior suggest that institutional isomorphism may occur within established institutional orders, but it takes significant external shocks to change core beliefs.14 According to Hall, repeated failure may move political decision-makers from firstorder learning through second-order learning to third-order learning, involving a reorientation in “the hierarchy of goals and set of instruments employed to guide policy.”15 Furthermore, while the conditions for policy change outlined above are necessary, they are not sufficient. Neither a crisis nor institutional receptivity automatically triggers a change in beliefs or policy transfer. Institutional learning and change are also filtered by policy-makers’ beliefs, and more particularly by coalitions’ conflicting beliefs.16 Our analysis here draws on insights from Sabatier’s advocacy coalition framework (ACF). The ACF is an actor-based framework developed for analyzing policy change. It focuses on ideologically-based advocacy coalitions, which consist of actors who are united by their beliefs and compete for control of policy domains. Sabatier identifies two kinds of beliefs: secondary beliefs that are likely to change over time, and normative or empirical core beliefs that are more stable.17 Sabatier worked out a theoretical model of policy change that combines a conflict resolution (power) approach with a learning approach. He argues that what is learned depends on the power of coalitions, but cannot be solely understood from the (changing) division of power. The ACF assumes therefore that while pre-existing policy beliefs act as a filter for their perception of new information, policy beliefs can slowly change via learning and interaction. Change is therefore analyzed not as a direct result of learning pressures or as a reaction to exogenous shocks disrupting the previous domestic equilibrium, but rather as the result of a regulated, institutionallyshaped process of negotiation and compromise between incumbent actors and others who use EU proposals to challenge existing welfare regimes. C. A Typology of Institutional Learning Since institutional change may take various forms, it is useful to develop a typology in order to better describe the asymmetric EMU progression over the past decade until the crisis. This article will draw on the typology developed by Streeck and Thelen, which delineates four types of institutional shift. In their schema, layering takes place when new institutional elements are added to existing ones; displacement is the process by which an institution increases its power; redirection is the modification of the parameters and mandates of an institution, whether marginal or fundamental; and drifting occurs when institutions are overwhelmed by external P. T. Wong & B. Weiner, When People Ask “Why” Questions, and the Heuristics of Attributional Search, 40 J. PERS. SOC. PSYCHOL. 650–663 (1981). 13 M. Popper & R. Lipshitz, Organizational Learning: Mechanisms, Culture, and Feasibility, 31 MANAG. LEARN. 181–196 (2000). 14 M. DIERKES ET AL., HANDBOOK OF ORGANIZATIONAL LEARNING AND KNOWLEDGE (2003). 15 P. A. Hall, Policy Paradigms, Social Learning, and the State: The Case of Economic Policymaking in Britain, 25 COMP. POLIT. 275–296 (1993). 16 P. A. SABATIER & H. C. JENKINS-SMITH, POLICY CHANGE AND LEARNING: AN ADVOCACY COALITION APPROACH (1993). 17 C. M. Weible & P. A. Sabatier, Coalitions, Science, and Belief Change: Comparing Adversarial and Collaborative Policy Subsystems, 37 POLICY STUD. J. 195–212 (2009). 12

6 developments. Based on this typology, the next sections discuss how EMU institutions have been progressively transformed in light of the Maastricht lock-in effect, and how the EMU path-dependent limitations have turned the GFC into a sovereign debt crisis. Specifically, the next section is concerned with the reinforcement of the existing fiscal and monetary institutions – through redirection, displacement and layering – and the “drifting” of social and employment institutions.

7

II. EXPLAINING THE EMU ASYMMETRIC LEGAL CONSTRUCTION: AN INSTITUTIONALIST ANALYSIS Using the typology described in Section I, this section aims to reconstruct the institutional processes that have (re)shaped the EMU in light of the Maastricht lock-in effect. Specifically, it is argued that fiscal, monetary and social institutions have evolved through a process of layering, redirection, displacement and drifting due to institutional friction and spillover effects stemming from the EMU’s asymmetric structure.18 This transformation process began with the Jacques Delors report, which attempted to place more weight on the “E” of EMU.” The idea was to create “two integral parts of a single whole” that would be implemented in parallel.19 In order to understand the institutional path-dependent processes associated with the EMU, we must first examine the theoretical legal foundations upon which the EMU project is based (A). Accordingly, we examine the extent to which the monetarist-asymmetric legal framework illustrates a process of institutional redirection and displacement of monetary and fiscal institutions (B). Next, we study the asymmetry of the EMU policy mix, which is characterized by decentralized economic and social policies (C). Thirdly, we explore the layering process engaged to reinforce social and employment policy learning and coordination in the hope of addressing EMU asymmetry (D). Finally, we consider whether this weak layering process constitutes drifting (E). A. Theoretical Tensions Underlying EMU Governance There is no grand structural theory that encompasses the entire European integration process, but over recent decades its economic constitutional framework has been influenced by two underlying paradigms. The first, neoliberalism,20 and its German variant, ordoliberalism, argue that “markets always perform optimally” and that public policy and law disturb “well-functioning markets” (1).21 The alternative model is based on a set of ideas that has many sources. The main strand, developed by John Maynard Keynes, argues that markets do not always work well, nor do they selfcorrect (2).22 1. Theoretical Redirection of EMU Governance: From Keynesianism to Neoliberalism In the post-war period it was widely accepted that markets ought to be subject to various forms of political control on the Keynesian model of macro-economic

18

M. SALINES ET AL., BEYOND THE ECONOMICS OF THE EURO: ANALYSING THE INSTITUTIONAL EVOLUTION OF EMU 1999-2010 42 (2011), http://www.ecb.europa.eu/pub/pdf/scpops/ecbocp127.pdf. 19 European Commission - Phase 3: The Delors Report Economic and Monetary Union (2010), Oct. 30, 2010. 20 The term ‘neoliberalism’ was originally coined in 1938 by the German scholar Alexander Rüstow at the Colloque Walter Lippmann. The concept draws on different school of thought (Freiburg school, the Austrian school, the Chicago school of economics, and Lippmann’s realism). See P. MIROWSKI & D. PLEHWE, THE ROAD FROM MONT PÈLERIN: THE MAKING OF THE NEOLIBERAL THOUGHT COLLECTIVE (2009). 21 R. Schettkat, Will only an Earthquake Shake Up Economics?, 149 INT. LABOUR REV. 185–207 (2010). 22 J. M. KEYNES, THE GENERAL THEORY OF EMPLOYMENT, INTEREST, AND MONEY (1965).

8 management.23 One of the problems that Keynes recognized in the American crisis of the 1930s was that wages could be too flexible. When wages fall, individual and household incomes also fall, as does consumer demand. Imposing more wage flexibility can then exacerbate the underlying problem of lack of aggregate demand. Accordingly, the Keynesian model assigns a primary function to fiscal policy, which is supposed to expand aggregate demand through tax cuts and deficit-financed expenditures in times of recession. Monetary policy plays only an accommodating role, namely the financing of expansionary policy at low interest rates in order to prevent the collapse of domestic demand during fiscal retrenchment. This model fell apart during the “Great Inflation” of the 1970s. 24 Specifically, the oil crisis of the 1970s collapsed the Bretton Woods systems and plunged the Keynesian compromise into a structural crisis. The Bretton Woods system of 1944, which was comprised of a harmonious trio of autonomous monetary policies, fixed exchange rates and increasing liberalization of international trade, became increasingly incompatible with the globalization of financial markets and spread of new technology.25 Market dynamics completely reversed and engaged in a paradigmatic redirection that had a significant impact on the interaction between social policies and the laws of economics. While this article does not try to give a full account of neoliberalism, it is worth noting that the concept has several definitions – classical, economic, philosophical and corrupted – as its meaning has changed over time, ultimately coming to mean different things to different groups. Furthermore, the most prominent authors on neoliberalism – Friedrich Hayek,26 Milton Friedman,27 David Harvey28 and Noam Chomsky29 – do not agree on the meaning of neoliberalism, and this lack of agreement presents significant obstacles to creating an unbiased and unambiguous definition. David Harvey’s definition sheds some light on the main foundational ideas of the concept: “Neoliberalism is in the first instance a theory of political economic practices that proposes that human well-being can best be advanced by liberating individual entrepreneurial freedoms and skills within an institutional framework characterised by strong private property rights, free markets and free trade. The role of the state is to create and preserve an institutional framework appropriate to such practices. The state has to guarantee, for example, the quality and integrity of money. It must also set up those military, defence, police and legal structures and functions required to secure private 23

Id. The European ‘Great Inflation’ of the 1970s was a period characterized by ‘stagflation’ – the simultaneous rise of inflation and unemployment. 25 The flow of capital in circulation ‘became simply too high to continue to coexist with a regime of fixed exchange rates and monetary sovereignty’. See K. R. McNamara, Consensus and Constraint: Ideas and Capital Mobility in European Monetary Integration, 37 JCMS J. COMMON MARK. STUD. 455–476 (1999); also, for more details on the decline of Keynesianism see T. Palley, From Keynesianism to Neoliberalism: Shifting Paradigms in Economics, in NEOLIBERALISM: A CRITICAL READER (Alfredo Saad-Filho ed., 2005). 26 F. A. HAYEK, THE CONSTITUTION OF LIBERTY (2013). 27 M. FRIEDMAN & A.J. SCHWARTZ, A MONETARY HISTORY OF THE UNITED STATES: 1867-1960 (1971). 28 D. HARVEY, A BRIEF HISTORY OF NEOLIBERALISM (2005). 29 R. W. McChesney, Noam Chomsky and the Struggle Against Neoliberalism, 50 MON. REV. INDEP. SOCIAL. MAG. 40–47 (1999). 24

9 property rights and to guarantee, by force if need be, the proper functioning of markets. Furthermore, if markets do not exist (in areas such as land, water, education, health care, social security, or environmental pollution) then they must be created, by state action if necessary. But beyond these tasks the state should not venture. State interventions in markets (once created) must be kept to a bare minimum because, according to the theory, the state cannot possibly possess enough information to second-guess market signals (prices) and because powerful interest groups will inevitably distort and bias state interventions (particularly in democracies) for their own benefit.”30 This neoliberal shift has directly affected the governance of employment policy, notably with the development of the income distribution and natural rate theories. According to the former, the market ensures that factors of production are paid what they are worth, obviating the need for institutions of social protection and trade unions. In practice this has taken the form of deregulatory pressure on the labor market – albeit with a lower impact in Europe, where the Keynesian model of redistributive economic policy held sway at the Member State level. 31 However, this resulted in falling real-minimum wages, undermining unions and employment security in many industrialized economies. As for the natural rate theory, it was developed by Milton Friedman, monetarism’s leading proponent. In his view, government intervention should be minimal and the money supply (the total amount of money in an economy, in the form of coin, currency and bank deposits) is key to fighting the effects of inflation. It argues that excessive expansion of the money supply is inherently inflationary and that monetary authorities should focus solely on maintaining price stability.32 In 1968, Friedman published an influential paper in which he argued that policy-makers should discard any Keynesian prescriptions of activist demand management – for instance, expansionary fiscal policy and excessive expansion of money supply – because these measures could (at best) reduce unemployment under the natural rate in the short run, and in the long run would only cause inflation. According to Friedman, the economy would always return to the natural rate of unemployment, that is, the lowest rate of unemployment at which inflation remains stable.33 In the 1980s, policy-makers sought to follow strict Chicago School monetarist prescriptions and abandoned Keynesian interest-rate fine-tuning in favor of moneysupply targeting. As a result, unemployment in Europe began to rise and persisted at high levels after each recession without returning to the pre-recession so-called equilibrium. Natural rate economists rejected the idea that restrictive macroeconomic policies were the cause of unemployment, focusing instead on the role of market institutions.34 They claimed that generous unemployment benefits, aggressive unions, strict employment protection laws and a more compressed wage structure were among the causes of less well-functioning labor markets.

30

HARVEY, supra note 29. Deakin, supra note 8. 32 M. Friedman and A.J. Schwartz, A Monetary History of the United States: 1867-1960 (Princeton University Press 1971). 33 M. Friedman, The Role of Monetary Policy, 58 AM. ECON. REV. 1–17 (1968). 34 R. M. Solow, Broadening the Discussion of Macroeconomic Policy, in ECONOMIC POLICY PROPOSALS FOR GERMANY AND EUROPE 20–28 (R. Schettkat & J. Langkau eds., 2008). 31

10 2. A Multi-Layered EMU: Ordoliberal Europe and the Emergence of a Subordinated Social Dimension In Europe, despite a robust national Keynesian resistance, the neoliberal redirection has profoundly altered the course of the European economic integration process.35 This shift did not take the form of a natural linear and/or mono-causal process; rather, it was the result of a layering process. It is argued that the so-called German “ordoliberal”36 model of legally constituted order serves as one of the main influences on EU legal thinking, although the extent of that influence remains unclear.37 According to the ordoliberal conception of the state-market relationship, law and politics have the task of establishing the conditions for a system of undistorted competition.38 Free competition is the predominant objective that can be achieved only through a pre-established economic order by means of an “economic constitution.”39 The constitutional framework must go beyond the mere enforcement of private law, property and contractual rights;40 it also has to guarantee free competition by regulating cartels and monopolies. At the same time, discretionary public intervention in the market and state ownership of industry have to be constrained in order to preserve competition.41 This view has significantly influenced the EU integration dynamics. As argued by Christian Joerges, “the fact that Europe had started its integrationist path as a mere economic community lent plausibility to ordoliberal arguments.”42 The provisions regarding competition and freedom of movement in the Treaty of Rome, the concept of “distortion of competition” mentioned in the Spaak Report on which the Treaty is based, and the freedoms guaranteed in the EEC Treaty43 can all be “interpreted as a ‘decision’ supporting an economic constitution that matched ordoliberal conceptions of the framework for a market economy system.”44 Against this background, several proponents of the ordoliberal theories realized that the market was not the natural order and thus argued for a different economic legal framework, one that could also achieve non-economic goals, including social policy. The asymmetry created by the ordoliberal compromise had to be addressed via a new approach to market regulation: the social market economy. According to the 35

C. Hermann & I. Hofbauer, The European Social Model: Between Competitive Modernisation and Neoliberal Resistance, 31 CAP. CL. 125–139 (2007). 36 The theory was developed in the 1930s by German economists and legal scholars from the Freiburg School such as Walter Eucken, Franz Böhm, Hans Grossmann-Doerth, Leonhard Miksch; see V. J. Vanberg, The Freiburg School: Walter Eucken and Ordoliberalism, The Freiburg School, Freiburg Discussion Papers on Constitutional Economics No 04/11 (Apr. 2004); F. Böhm, W. Eucken & H. Grossmann-Doerth, The Ordo Manifesto of 1936, in GERMANY’S SOCIAL MARKET ECONOMY 15– 26 (A. Peacock & H. Willgerodt eds., 1989). 37 C. Joerges, What is Left of the European Economic Constitution? A Melancholic Eulogy, 30 EUR. LAW REV. 461 (2005); D. J. Gerber, Constitutionalizing the Economy: German Neo-Liberalism, Competition Law and the “New” Europe, 42 AM. J. COMP. LAW 25–84 (1994). 38 Deakin, supra note 8. 39 H. J. Blanke, The Economic Constitution of the European Union, EUR. UNION LISBON 369–419 (2012). 40 Deakin, supra note 8. 41 C. JOERGES & F. RÖDL, “SOCIAL MARKET ECONOMY” AS EUROPE’S SOCIAL MODEL? (2004). 42 Id. at 13. 43 Namely the opening of national economies, anti-discrimination rules, and the commitment to a system of undistorted competition. 44 JOERGES AND RÖDL, supra note 42.

11 German economist Alfred Muller-Armack, the social market economy is aimed at finding a third way between pure “laissez faire” and “planned economy.” 45 The concept of social market economy remains significantly interwoven with ordoliberalism, but it also aims to achieve social justice and social protection of individuals.46 The idea was to combine “more socialism with more freedom”47 and to combine “the principle of market freedom with the principle of social balance.” The essence of social market economy is to ensure societal wellbeing via undistorted competition and economic growth. Social market philosophy does not shift away from the ordoliberal emphasis on competition: it only redefines competition as an instrument for social achievements. Furthermore, Muller argues that relying on the social benefits generated by the market is not sufficient. Sustainable growth and social cohesion require protection of the “elements of social fabric”48 and extension of the function of the State during times of economic downturn and market recession. State interventionism should, however, be constrained such that it operates within a strictly delineated and liberal framework.49 The Lisbon Treaty makes explicit reference to a social market economy as a policy goal.50 One may nonetheless question whether this ambiguous notion can effectively serve as a constitutional objective. Notwithstanding the commendable desire of the Treaty’s drafters to develop a social balance, the concept remains influenced by neoliberalist and ordoliberalist thinking, and so the goal of achieving a genuine social Europe would seem to be unattainable. Without clearer EU objectives in the field of social policy, the concept of a social market economy can only bring the national Keynesian models of redistributive economic policy and European ordoliberalism into greater conflict. B. An Asymmetric-Monetarist Legal Framework: Monetary Displacement and Fiscal Redirection The following section will examine how the monetary and fiscal institutions of the EMU have evolved through a process of displacement and redirection. More precisely, we will discuss how the contested monetarist rationale behind Maastricht has been imposed and displaced via legal means (1) and examine the extent to which fiscal policy has been redirected (2). 1. Displacement of a Contested Version of the EMU a. Monetary Obligations The discourse of Maastricht called for a first legal obligation: “price stability.” This principle, as laid down in Article 3a of the Maastricht Treaty (Article 119 45 J. SCHMIDT, ALFRED MÜLLER-ARMACKS BEITRAG ZUR THEORIE, PRAXIS UND REFORM DER SOZIALEN MARKTWIRTSCHAFT (2007). 46 W. SAUTER, COMPETITION LAW AND INDUSTRIAL POLICY IN THE EU (1998). 47 A. MÜLLER-ARMACK, GENEALOGIE DER SOZIALEN MARKTWIRTSCHAFT 46 (1974). 48 Deakin, supra note 8 at 23. 49 Labelled ‘liberal interventionism’. See MÜLLER-ARMACK, supra note 48. 50 See Article 3 TFEU: ‘The Union shall establish an internal market. It shall work for the sustainable development of Europe based on balanced economic growth and price stability, a highly competitive social market economy, aiming at full employment and social progress, and a high level of protection and improvement of the quality of the environment’.

12 TFEU), constituted the legal template for EMU macroeconomic management. Price stability was defined in the Treaty as “the primary objective” of the European System of Central Banks (ESCB) and foreshadowed the establishment of a new and independent ECB that would design and implement a single monetary policy to that effect. A conventional argument in favor of this model claimed that a stable currency among participant countries removed the basic uncertainty hindering the deepening of a single market. It was expected that transaction costs would simultaneously be reduced, enhancing both external trade between Member States and the profitability of firms. It was also a commonly held view that investors sought low inflation51 and preservation of the value of the currencies in which bonds were denominated.52 To this end, monetary policy had to be removed entirely from the political process. 53 This involved the establishment of independent central banks that could pursue a low- or zero-inflation target, with rules to prohibit government deficits through debt monetization.54 b. Displacement of Monetary Institutions This version of the EMU has been a contested topic since its creation. There was never a clear-cut break between its Keynesian and neoliberal phases. However, there was a significant intellectual shift and displacement between the Werner Plan and the Maastricht Treaty. The Werner Plan was aimed at an Economic and Monetary Union, while Maastricht is restricted to a Monetary Union. The “economic dimension” (as opposed to the monetary dimension) was not as well developed as in the Werner Plan.55 The political economy of the Werner Plan referred to a polity that emphasized social justice through political distribution, but only as long as market relations were regulated. This does not mean that the market controls the EMU, nor does it mean that the prospect of Social Europe is gone forever; this is a misconception of the foundations of the EMU. Several EMU founders had in fact clearly understood that the market is not a natural force, not even in its most extreme laissez faire version. However, in spite of this social rationale, it is difficult to hope for a resurgence of the Werner rationale, particularly in the context of the eurozone crisis. EMU normative goals laid down in Maastricht remain limited to monetary stability, stable currency rates, fiscal discipline and, eventually, a low rate of inflation. The Maastricht Treaty does not comprise a comprehensive spectrum of economic policy, including social and labor. In spite of the controversial nature of this model, law played a determining role in allowing the displacement and reinforcement of this contested version of the EMU. Indeed, juridification seems to have played a critical role in supporting this controversial conception of the Monetary Union. What we critique in this paper is not the ordoliberal norms underlying the EMU, although their theoretical and empirical grounding have always been contested by a majority of non-monetarist economists. 51

According to neoclassical economists, high inflation has a negative impact on growth rates. Hence the term "sound money." 53 A. Alesina & R. Gatti, Independent Central Banks: Low Inflation at No Cost?, 85 AM. ECON. REV. 196–200 (1995). 54 Monetarization involves the purchase of government debt by the central bank, which in turn increases the money supply. 55 L. MAGNUSSON & B. STRATH, FROM THE WERNER PLAN TO THE EMU: IN SEARCH OF A POLITICAL ECONOMY FOR EUROPE 1 (2001). 52

13 The problem plaguing Maastricht is not (only) its ordoliberal rationale, but also the fact that legal maneuvering facilitated the construction of a controversial conception of the EMU.56 2. Incremental Redirection of Fiscal Policy a. Fiscal Obligations The second legal objective associated with the creation of the EMU was budgetary stability: receipts and expenditures should more or less match each other, and any deficits should be limited, short-lived and hence sustainable. The underlying idea was that the euro had to be seen by the international finance community as viable and legitimate. To build sufficient credibility, Member States had to maintain low and stable ratios of debt to GDP.57 Accordingly, Article 104 EC (Article 126 TFEU) introduced a disciplinary legal mechanism, the so-called Excessive Deficit Procedure (EDP). The Commission defines the EDP as “an action launched by the European Commission against any European Union (EU) Member State that exceeds the budgetary deficit ceiling imposed by the EU’s Stability and Growth Pact legislation. The procedure entails several steps, potentially culminating in sanctions, to encourage a Member State to get its budget deficit under control, a requirement for the smooth functioning of Economic and Monetary Union (EMU).”58 According to the “Protocol on the Excessive Deficit Procedure,” annexed to Maastricht, the EDP was intended to monitor the state of convergence with respect to specific fiscal criteria according to which the government deficit must not exceed 3% of GDP and government debt must not exceed 60% of GDP.59 Moreover, the monitoring of budgetary policies was to be ensured by the Commission, which could recommend that the Council take disciplinary measures in case of non-compliance with the fiscal criteria. These sanctions could take the form of non-interest-bearing deposits or fines (Article 104(11) EC). The Maastricht fiscal pillar was, however, considered insufficient to guarantee the smooth introduction of the euro. The need to reinforce discipline once inside the EMU was recognized by the Madrid European Council in December 1995 and reiterated in Florence six months later. Accordingly, the Amsterdam Treaty of 1997 further reinforced the fiscal pillar of the EMU through a new fiscal agreement, the Stability and Growth Pact (SGP).60 The principal objective of the SGP was to safeguard sound government finance in order to strengthen the conditions for price stability, which would in turn result in strong growth and employment creation. It was recognized that the lack of exchange rate flexibility would indicate a greater role for automatic fiscal stabilizers in 56 57

JOERGES AND RÖDL, supra note 42. M. BUTI, MONETARY AND FISCAL POLICIES IN EMU: INTERACTIONS AND COORDINATION

(2003). 58

Eurostat, Excessive Deficit Procedure (EDP) - Statistics Explained Glossary, Aug. 5, 2013. A potentially confusing peculiarity of the EDP is that the word ‘deficit’ is used to refer both to situations where either the deficit or the debt is too high. In some cases, where the procedure is different for deficit and debt, it will be specified clearly. See European Commission, The Corrective Arm - EU Economic Governance, Sept. 5, 2013. 59 Protocol on the convergence criteria referred to in Article 109j of the Treaty establishing the European Community, 1992 O.J. (C191). 60 Entered into force on January 1, 1999.

14 adjusting to asymmetric shocks, which would require the Union to “ensure that national budgetary policies support stability-oriented monetary policies.”61 According to this rationale, the SGP would allow all “Member States to deal with normal cyclical fluctuations while keeping the government deficit within the reference value of 3% of GDP.”62 In sum, the objective was to restrict the use of automatic stabilizers. b. Redirection of Fiscal Institutions after 2005 The evolution of fiscal policy provides useful insights into the process of institutional redirection. The SGP, which was originally a disciplinary device, became more flexible in 2005 following an intense legal debate before the European Court of Justice. It was argued that the disciplinary mechanisms associated with the SGP increased tensions within the EMU and robbed the legal outcome of any legitimacy and effectiveness. Both the German Constitutional Court and the European Court of Justice were confronted with this issue in the matter of the SGP implementation. Firstly, in spite of the constitutional barriers to the implementation of the SGP, the German Court in its Maastricht judgment63 regarding the constitutionality of the SGP and argued that the Law had endowed the Monetary Union with a democratic political structure of its own. The viability of this constitutional reasoning proved, however, to be rather limited when Germany (along with others countries including France and the Netherlands) failed to respect the Maastricht Treaty, which led the Commission to take legal action before the ECJ. However, contrary to the German Court’s ruling, the European Court took a modest approach in dealing with the violation of the SGP.64 It sidestepped key questions on the nature and the sequence of the violation65 and did not engage in a legal dispute with Member States’ sensitive interpretation of the SGP criteria.66 In spite of the hard legal backing accompanying the SGP, it seemed wise to disregard the obvious violation, given the lack of the political and legal support for the SGP at the time. As argued by Joerges, the law could not “substitute for the necessary historical evolution of equally Europeanized social preconditions for successful monetary operation.”67 He claims that policy-makers should not have forced the juridification of EMU without genuine political bargaining. Doing so created an unforgiving monetary policy operating in a vacuum and prevented any adjustment for socio-economic disparities within the Union.68 Accordingly, the SGP was reforms of 2005 redirected its logic in two ways. First, 61

Id. Id. 63 Judgment on the Maastricht Treaty of October 12, 1993, Cases 2 BvR 2134/92, 2 BvR 2159/92 (Bundesverfassungsgericht – BVerfG), reported in 89 Official Court Reports 155 [BVerfGE 89, 155], Translation by Wegen et al., 33 I.L.M. 388 (March, 1994). 64 Commission of the European Communities v Council of the European Union, Case C-27/04 (Euro. Ct. of Justice (Apr. 24, 2004). 65 B. Dutzler & A. Hable, The European Court of Justice and the Stability and Growth Pact-Just the Beginning?, 9 EUR. INTEGR. ONLINE PAP. EIOP 20 (2005). 66 For a more detailed analysis of the judgment see Joerges, supra note 38. 67 M. Everson & C. Joerges, Reconfiguring the Politics–Law Relationship in the Integration Project through Conflicts–Law Constitutionalism, 18 EUR. LAW J. 35 (2012). 68 M. Herdegen, Price Stability and Budgetary Restraints in the Economic and Monetary Union: Law as the Guardian of Economic Wisdom, 35 COMMON MARK. LAW REV. 9–32 (1998). 62

15 it extended budgetary surveillance horizontally by focusing more on the long-term perspective and on fiscal “sustainability.” The debt criterion became more important than the deficit.69 Second, it extended surveillance vertically by looking at the composition of public finance in the Member States’ budgets and by adapting the SGP to national conditions with country-specific medium term objectives.70 C. Economic and Social Deficit of the EMU As stated by Pascal Lamy,71 “we called economic and monetary union a union that was extremely monetary and scarcely economic.”72 The European integration process was supposed to be sequenced as follows: custom union, single market, economic and monetary union, and finally political union, albeit with a great vagueness concerning the economic (1) and social dimensions (2). 1. Coordination of Economic Policy While monetary policy is a competence of an outright federal nature, there was no equivalent supranational decision-making process in the field of economic policy. Member States retained control of the means to the ends in that Member States and the Union shared competence over economic policy, with the SGP shaping it in accordance with convergence criteria. Member States were only required to regard their economic policies as a matter of common concern and coordinate them with the Council, which had the task of issuing broad guidelines for these policies and was responsible for the multilateral surveillance procedure that was provided to ensure proper compliance with those guidelines.73 However, as coordination procedures tended to be subordinated to the primary objective of monetary stability and became more stringent, soft law gradually gave way to hard law. Most important provisions regarding economic policy coordination are drawn from the Maastricht Treaty. Articles 98-104 EC (Articles 121-126 TFEU) provide the core principles and standards by which to achieve a coordinated economic policy. According to Article 4(1) EC (Article 119 TFEU), the Union and Member States’ economic policies should be based on the “close coordination of Member States’ economic policies, on the internal market and on the definition of common objectives, and conducted in accordance with the principle of an open market economy with free competition.” The main provision regarding economic policy is Article 99 EC (Article 121 TFEU), which concerns Broad Economic Policy Guidelines (BEPG) and multilateral surveillance. According to 121(1) TFEU, Member States shall regard their economic policies as a matter of common concern and shall coordinate them with the Council.

69

European Commission, Communication from the Commission to the Council and the European Parliament - Public Finances In EMU, 2006. COM(2006)304 final. 70 C. JOERGES & F. RÖDL, “SOCIAL MARKET ECONOMY” AS EUROPE’S SOCIAL MODEL? (2004). 71 Current president of the WTO and former Chief of Staff of Jacques Delors. 72 The French version reads as follows: ‘on an appelé union économique et monétaire une union qui était en réalité très monétaire et guère économique.’ See P. Lamy & H. Busson, Pourquoi l’Euro ?, 11 REGARDS CROISÉS SUR ECON. 12–22 (2012).. 73 Article 121 TFEU (ex Article 99 TEC).

16 2. The Limited Social Dimension of Maastricht On February 7, 1992, the signatories to Maastricht adopted a Protocol on Social Policy and an Agreement, annexed to the Protocol, between eleven Member States (with the exception of the United Kingdom.)74 The Social Policy Protocol was a legal mechanism adopted to overcome the deadlock reached over the social provisions of Maastricht at the summit of December 1991. Another important substantive contribution was the extension of qualified majority voting in the field of social policy. The Protocol attenuated national sovereignty and created a new means of overcoming deadlock on a number of measures that had been stalled by the United Kingdom. A significant contribution of the Agreement on Social Policy was the constitutional recognition of a role for the social partners in the Community legislative process.75 The lack of development in the social field strengthened the single-sided monetarist approach underlying the EMU construct. Besides the Social Policy Protocol, Maastricht did not introduce any form of social bulwark that could counterbalance financial and monetary obligations – not even broadly comparable with the ESCB before the launch of EMU. There were, for instance, no community stabilization and/or redistribution mechanisms to protect against asymmetric shocks. For some, the social dimension was being “structurally and consciously underdeveloped” vis-à-vis monetary and economic aspects in order to give way to market forces that would eventually restructure the “expensive” welfare states.76 3. Alleged Downward Pressures on Social and Labor Systems Whereas the drafters of the Maastricht Treaty predicted that the EMU would generate social spillovers, others claimed that the incomplete and restrictive nature of the EMU governance model would yield disappointing results for national welfare.77 In particular, the stringent budgetary discipline combined with uniform monetary policy was said to introduce a downward learning pressure on national “Keynesian capacity”78 along with a deregulatory version of “labor market flexibility.”79 74

The UK ratified the Protocol not long afterward. The new Social Chapter of the EC Treaty, as amended by the Treaties of Amsterdam and Nice (Articles 136 EC, now Articles 151 TFEU)), incorporates the new Social Chapter of the EC Treaty, as amended by the Treaties of Amsterdam and Nice (Articles 136 ff. EC, now Articles 151 TFEU)), incorporates the Agreement on Social Policy. 76 A. Verdun, An “Asymmetrical” Economic and Monetary Union in the EU: Perceptions of Monetary Authorities and Social Partners, 20 J. EUR. INTEGR. 59–81 (1996); P. POCHET & B. VANHERCKE, SOCIAL CHALLENGES OF ECONOMIC AND MONETARY UNION (1998); C. De la Porte & B. Vanhercke, Still Building Social Europe through the Open Method of Coordination?, in ANOTHER EUROPE IS POSSIBLE? ALTERNATIVES AND RESISTANCE TO NEOLIBERAL EUROPEAN GOVERNANCE. PANEL LABOUR IN THE PROCESS OF RESTRUCTURING AND LIBERALISATION OF EUROPEAN GOVERNANCE (2010). 77 M. Rhodes, Globalization and West European Welfare States: A Critical Review of Recent Debates, 6 J. EUR. SOC. POLICY 305–327 (1996); S. Gill, European Governance and New Constitutionalism: Economic and Monetary Union and Alternatives to Disciplinary Neoliberalism in Europe, 3 NEW POLIT. ECON. 5–26 (1998); P. van den Bempt, The Political Aspects of Economic and Monetary Union: A View from Brussels, in THE POLITICS OF ECONOMIC AND MONETARY UNION 19–54 (P. Minkkinen & H. Patomäki eds., 1997). 78 W. Streeck, Neo-Voluntarism: A New European Social Policy Regime?, 1 EUR. LAW J. 31–59 (1995). 79 S. Deakin & H. Reed, The Meaning of Labour Market Flexibility: Economic Theory and the 75

17 Many analysts argued that given the high level of budget deficits and public debts, the stringent EMU framework would necessitate radical fiscal retrenchment and trigger a race to the bottom.80 Countries with flexible welfare systems would enjoy an artificial competitive advantage over those with more extensive welfare provisions. As a result, producers in high social-protection jurisdictions would be at a disadvantage vis-à-vis their competitors with lower social standards. Furthermore, in order not to boost competitiveness, strong welfare systems would have to dismantle their social protection, leading to an unraveling of the national social fabric. The Maastricht macroeconomic framework was also said to exert significant pressure on labor systems.81 It was argued that labor market would become a key substitute for nominal exchange rate fluctuations and expansionary fiscal policy (which is not possible under a single currency.) As Member States lost the “easy option” of adjusting to changing economic conditions through devaluation and public spending, economic shocks had to be tackled by means of supply-side measures, thus making markets and industries operate more efficiently. It was argued that the labor market had to operate with maximum “flexibility” – that is, flexibility in the sense of lowering wages and labor standards to bring them into line with cost competitiveness requirements.82 The concept of labor flexibility has many meanings, and deregulation is not an inherent component of increased flexibility. For instance, Deakin and Reed envisage several versions of labor market flexibility that do not necessitate a trade-off between regulation and flexibility.83 On the contrary, there is a strong case that labor law in many instances enhances long-term growth and competitiveness. However, this definition of the concept did not work into practice. The policy discourse of the Union since the 1990s seems clearly entrenched in a deregulatory definition of flexibility, which is said to be the “corollary to the process of EMU.”84 Several legal and policy documents associated with Maastricht stress the importance of structural reforms as means of eliminating “rigidities” within the labor market. This includes reforms of employment protection legislation and the shifting of the tax burden from employment to consumption. For instance, the European Council Resolution on Growth and Employment, adopted together with the SGP, clearly stated the need to “develop a skilled, trained and adaptable workforce and to make labour Discourse of European Integration, (Centre for Business Research, University of Cambridge, ESRC Working Papers No 163, 2000), available at http://core.ac.uk/download/pdf/7151278.pdf (last visited Jul 27, 2012). 80 S. Deakin & C. Barnard, Social Policy and Labour Market Regulation, in OXFORD HANDBOOK ON EUROPEAN UNION LAW (A. Menon, E. Jones, & S. Weatherill eds., 2013); S. Deakin & R. Rogowski, Reflexive Labour Law, Capabilities, and the Future of Social Europe, in TRANSFORMING EUROPEAN EMPLOYMENT POLICY: LABOUR MARKET TRANSITIONS AND THE PROMOTION OF CAPABILITY 229–255 (R. Rogowski, R. Salais, & N. Whiteside eds., 2011); R. ROGOWSKI, R. SALAIS & N. WHITESIDE, TRANSFORMING EUROPEAN EMPLOYMENT POLICY: LABOUR MARKET TRANSITIONS AND THE PROMOTION OF CAPABILITY (2012) at 229. 81 DEAKIN AND REED, supra note 80; S. GIUBBONI, SOCIAL RIGHTS AND MARKET FREEDOM IN THE EUROPEAN CONSTITUTION: A LABOUR LAW PERSPECTIVE 2 (2009); T. Peters, European Monetary Union and Labour Markets: What to Expect?, 134 INTL LAB REV 315–333 (1995); S. Giubboni, Social Rights and Market Freedom in the European Constitution: A Labour Law Perspective (Cambridge University Press 2009), chap. 2; T. Peters, ‘European Monetary Union and Labour Markets: What to expect?’ (1995) 134 Int’l Lab. Rev. 315–333. 82 DEAKIN AND REED, supra note 79. 83 For more a more precise definition of flexibility see ibid. 84 Id. 36.

18 markets responsive to economic change.”85 The resolution also called for a more “employment-friendly” tax and social protection system aimed at improving the functioning of the labor market. In addition, the broad economic policy guidelines (Article 121 TFEU) have constantly advanced flexibility of the labor market as a key component of economic policy aimed at achieving high employment.86 D. Addressing EMU Asymmetry Through Layering: The Legal Basis for Employment Policy After the creation of the currency union, several actors advocated a more balanced EMU framework. As a result, the Treaty of Amsterdam (Articles 145-150 TFEU) added several layers to the existing weak social institutions, namely the Broad Economic Policy Guidelines (BEPGs), which were put into place in 1993. The following section examines this layering process, and more specifically the reinforcement of the legal framework for employment policy, beginning with Amsterdam (1) and continuing with the more recent Europe 2020 strategy (2). 1.The OMC Template of Amsterdam: European Employment Strategy The most significant layer of innovation since the Maastricht Treaty was the European Employment Strategy, which was the first supranational method of coordination. According to Article 145 TFEU (ex 125 EC): “Member States and the Union shall, in accordance with this Title, work towards developing a coordinated strategy for employment and particularly for promoting a skilled, trained and adaptable workforce and labor markets responsive to economic change with a view to achieving the objectives defined in Article 3 of the Treaty on European Union.” Article 146 TFEU (ex 126 EC) requires the Member States to coordinate their policies for the promotion of employment within the Council in a way that is consistent with the Broad and Economic Guidelines (BEPG) laid down in the provisions for economic policy (Article 148(2) TFEU). Before the Amsterdam Treaty came into force, the European Council decided to put the relevant provisions regarding employment policy monitoring into effect. The outcome, which came to be known as the European Employment Strategy, was agreed upon at an extraordinary meeting of the European Council of Luxembourg in November 1997. 2. Lisbon: Confirmation of the EES/OMC a. Lisbon Employment Priorities: “Modernising the European Social Model” and Active Labor Market Policies The Lisbon summit, which was held on March 23-24, 2000, set a “new strategic goal” for the Union in order to “strengthen employment, economic reform and social

85 Resolution of the European Council on the Stability and Growth Pact Amsterdam, 17 June 1997, O.J (C236). 86 DEAKIN AND REED, supra note 79.

19 cohesion as part of a knowledge-based economy.”87 This strategic goal was for the Union to become “the most competitive and dynamic knowledge-based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion.”88 The Lisbon Conclusions had three main objectives: (1) preparing the transition to a knowledge-based economy; (2) modernizing the European social model; and (3) sustaining a healthy economic outlook and favorable growth prospects.89 The central objective of the Lisbon Strategy was to achieve full employment through a section titled “Modernising the European Social Model.” According to the Lisbon Conclusions, it was crucial to invest in people and to build an active welfare state to address the issue of unemployment.90 This approach was said to rely on the Nordic flexicurity model, which combines flexible labor markets and adequate security.91 Indeed, the Nordic approach combines open markets and job flexibility with all the support employers require to restructure their workforce to meet changing demands,92 and in Scandinavia this model has led to 50 years of economic growth, low inflation, and relatively low unemployment;93 thus, Lisbon has attempted to implement this model at the EU level. To this end, four objectives were defined: education and training; more and better jobs (although the emphasis was more on the quantity than the quality of jobs); modernizing social protection; and promoting social inclusion (discussed below). b. Institutionalization of a Generic OMC Process Although the learning process was formalized with the endorsement of the OMC procedure in 2000, the Lisbon summit merely gave a name to an already-existing process of governance, providing an opportunity for EU policy-makers to recast existing initiatives as examples of the OMC for social inclusion. Many areas – including employment, social inclusion and social protection – were not suddenly transformed by Lisbon. The primary inspiration for the OMC was the EES, which was formalized by the 1997 Treaty of Amsterdam. The EES was essentially a new layer on top of the “Essen” process begun in 1994, the impetus for which was the 1993 White Paper on Growth, Competitiveness and Employment.94 The method was described as a “means of spreading good practice and achieving greater convergence on the main EU goals through common targets and Guidelines for Member States, something backed up by national action plans.”95 It relies on Lisbon European Council, ‘Presidency Conclusions’ 23 and 24 March 2000. Id. para 5. 89 Id. para 5. 90 Id. para 24. 91 See Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions – Towards Common Principles of Flexicurity: More and better jobs through flexibility and security, Jun. 27, 2007. COM(2007) 359 final . 92 R. Rogowski, Governance of The European Social Model: The Case of Flexicurity, 43 INTERECONOMICS 82–91 (2008). 93 P. Emmenegger, The Long Road to Flexicurity: The Development of Job Security Regulations in Denmark and Sweden, 33 SCAND. POLIT. STUD. 271–294 (2010); U. Becker, The Scandinavian Model: Still an Example for Europe?, 2007 INT. POLIT. GES. 41–57 (2007). 94 C. Barnard & S. Deakin, A Year of Living Dangerously? EC Social Rights, Employment Policy, and EMU, 30 IND. RELAT. J. 355–372 (2003). 95 Communication from the Commission – European governance - A white paper, Jul. 25 2001. COM(2001) 428 final. 87 88

20 regular monitoring of progress to meet those targets, allowing Member States to compare their efforts and learn from the experience of others.96 In the field of employment policy, the “new” OMC has reorganized the EES as follows: 1. Agreeing on common objectives; 2. Establishing common indicators as a means of fostering mutual learning; 3. Drafting NAPs/NRPs; monitoring and evaluating NAPs; 4. Establishing a more specific program of cooperation and mutual learning. More generally, Lisbon confirmed the EES process and renamed it, but more importantly, it extended the methods to other areas. Terms such as soft law, self-regulation, and negotiated governance have been widely used to characterize the OMC and the modes of governance that resulted from Lisbon.97 The OMC attracted much of the scientific debate, as it represented an important break with the old community method; it was characterized by experimentation, knowledge creation and flexibility of normative and policy standards.98 In procedural terms, the Lisbon Strategy consisted of new forms of multilevel governance through exchange of information among policy-makers, learning from others’ practices and intentions, national ownership, and finally the exertion of peer pressure to galvanize governments into taking appropriate policy action. This multilateral coordination of employment policies was said to be an effective alternative to EMU monetarism in the sense it could provide some safeguards against the temptation of Member States to protect domestic jobs through “beggar-thyneighbor” policies, competitive deregulation and tax cuts.99 Indeed, the OMC was meant to represent a protective barrier against the most harmful forms of regulatory competition while simultaneously creating the premise for a reduction of the “institutional gap”100 between monetarist policies and employment objectives. However, the asymmetry in depth and weight between the powers involved in the different policies is still obvious:101 Employment policies are still subordinated to monetary policy. 3. Employment Policy under the Europe 2020 Strategy The Lisbon Strategy was reformed in 2005 following a critical report of the High Level Group chaired by former Dutch Prime Minister Wim Kok.102 According to the Report the Lisbon Strategy was insufficiently focused, as it was “about everything 96

Id. G. FALKNER, COMPLYING WITH EUROPE: EU HARMONISATION AND SOFT LAW IN THE MEMBER STATES (2005). 98 J. Scott & D. M. Trubek, Mind the Gap: Law and New Approaches to Governance in the European Union, 8 EUR. LAW J. 1–18 (2002). 99 F. W. Scharpf, ‘Introduction: the Problem-Solving Capacity of Multi-level Governance’ (1997) 4 Journal of European Public Policy 520, 159. 100 F. G. Snyder, ‘EMU Revisited: Are We Making a Constitution?: What Constitution are we Making?’ in G. De Búrca and P. P. Craig (eds), European Union Law: An Evolutionary Perspective (Oxford University Press 1999); S. Giubboni, Social Rights and Market Freedom in the European Constitution: A Labour Law Perspective (Cambridge University Press 2009), 122. 101 S. Giubboni, Social Rights and Market Freedom in the European Constitution: A Labour Law Perspective (Cambridge University Press 2009), 123. 102 W. Kok, Facing the Challenge: The Lisbon Strategy for Growth and Employment -- Report from the High Level Group, Nov. 2004; the report was rejected but it triggered a revision of Lisbon. We will return to the Kok Report in Section II. This paragraph focuses only on Europe 2020. 97

21 and thus about nothing’, and ‘too many targets will be seriously missed.” In 2010 it was replaced by its successor, the Europe 2020 strategy. In spite of the mixed results and the economic crisis, Europe 2020 draws its inspiration in terms of both content and process from the Lisbon Strategy. The Europe 2020 Union strategy intends to create jobs and promote growth through economic and social reforms, while respecting environmental limitations. Under the three headings of smart, sustainable and inclusive growth, it covers policy actions at both national and EU levels and is aimed at enhancing the welfare of European citizens.1 This is to be achieved by pursuing the following five quantified headline targets: 1. Raising the employment rate for those aged 20-64 to 75%; 2. Raising combined public and private R&D investment to 3% of GDP; 3. Reducing greenhouse gas emission by 20% from 1990 levels; 4. Reducing school drop-out rates to less than 10%; and increasing the share of 30-34-years-olds having completed tertiary or equivalent education to at least 40%; 5. Reducing the number of people suffering or at risk of poverty and social exclusion by at least 20 million.1 The Europe 2020 Strategy is not explicit about the steering mechanisms to be used to implement the flagship initiatives. However, it seems to follow the Lisbon learning approach by using the OMC, although it has attempted to correct its weaknesses, mainly by giving the European Council a strong role in steering the implementation of the reform agenda. E. The Drifting of Social and Economic Institutions This section investigates the claim that EU social institutions are undergoing a drifting process insofar as Lisbon and Europe 2020 have been distorted by the neoliberal discourse of the EMU (1), thereby reaffirming the EMU asymmetry (2). We will then review the evidence and consider whether EU social policy illustrates a process of drifting (3). 1. The Distorted Rationale Behind Lisbon: Competiveness and Fiscal Bias Lisbon looked like the “quintessential utopia,”103 aimed at the attainment of growth, productivity, social inclusion and sustainable development. However, this ambitious agenda put forward contradictory socio-economic objectives hardly reconcilable in the EMU asymmetric context. It seems difficult indeed to modernize the European Social model, investing in people and combating social exclusion while at the same time applying a disciplinary macroeconomic policy mix.104 Moreover, the J. Creel, E. Laurent & J. Le Cacheux, Delegation in Inconsistency: the “Lisbon Strategy” Record as an Institutional Failure, (Observatoire Français des Conjonctures Economiques (OFCE), Sciences Po, Document No 2005-07, Jul. 2007) available at http://www.mzes.unimannheim.de/projekte/typo3/site/fileadmin/research%20groups/6/reader/Creel-LaurentLe%20Cacheux-Connex%202005.pdf (last visited Aug 1, 2014). 104 J. P. Fitoussi and L. Eloi, Union Monétaire et Modèle Social en Europe: Chronique d’une Incohérence Institutionnelle, paper presented at the Conférence de Recherche de l’Institut International 103

22 asymmetry in depth and weight between the powers involved in the different policies is still obvious. Many argued that employment and social policies have never been genuine priorities and are still subordinated to monetary policy. Lisbon and the OMC were said to be window-dressing, hiding an economic agenda regarding macroeconomic discipline and competiveness – an agenda aimed at dismantling social institutions. A first crucial problem with the Lisbon Strategy was its fiscal bias. Specifically, the Integrated Guidelines, the basis of a new EES process within the re-launched Lisbon, required Member States to submit annual SCPs to ensure the long-term sustainability of public finance.105 It was argued that counter-cyclical monetary and expansionary fiscal policies would reduce incentives for reform,106 and that profligate governments generally favor time-inconsistent and inflationary policies. Thus, removing counter-cyclical instruments from Member States’ economic arsenals was expected to foster (deregulatory) labor reforms.107 Secondly, Lisbon was also significantly influenced by the pursuit of competitiveness. The competitiveness rationale began to emerge with the publication of the White Paper on Growth, Competitiveness and Employment in 1993.108 The White Paper addressed the issue of the low employment rates prevailing in EU countries, unlike Japan, the United States and the then EFTA states. Instead of advocating a “quick fix” for the EU’s unemployment issues, it suggested a combination of macroeconomic and structural measures: firstly, greater flexibility in the economy as a whole, and in the regulatory framework in particular, which should become more enterprise-friendly; secondly, the creation of an efficient labor market able to respond to new competitive situations; thirdly, an open international environment.109 Following the bursting of the Asian IT bubble the Lisbon Strategy underwent a major strategic reappraisal best represented by the Kok Report (2004).110 It was argued that the main problem with Lisbon was its over-ambitious and contradictory approach. While social and employment policy were moved higher up the agenda, no consideration was given to the tensions between the quest for competitiveness and the idea of social balance.111 The Report therefore recommended a (re)focusing of priorities of the EES on boosting employment at all costs by making labor markets more flexible. Additionally, the Lisbon Strategy had to be geared to the paradigms of innovation, the internal market and administrative deregulation in order to promote economic growth and employment. In 2005, then, the EES was integrated into d’Etudes Sociales Sciences Po (Nov. 22, 2006). 105 See Council Recommendation of Nov. 19, 2009 on the 2009 update of the broad guidelines for the economic policies of the Member States and the Community and on the implementation of Member States’ employment policies, 2009 O.J. (L183/1). 106 D. Rodrik, Understanding Economic Policy Reform, 34 J. ECON. LIT. 9–41 (1996). 107 L. Calmfors, Fiscal Policy to Stabilise the Domestic Economy in the EMU: What Can We Learn from Monetary Policy?, 49 CESIFO ECON. STUD. 319–353 (2003). 108 European Commission, White Paper on Growth, Competitiveness and Employment - The Challenges and Ways Forward into the 21st Century. Dec. 5, 1993. COM (93) 700. 109 Id. 50. 110 Kok supra 102. 111 C. M. Radaelli, The Open Method of Coordination: A new Governance Architecture for the European Union?, Swedish Institute for European Policy Studies March, Working Paper, No 2003:1, Mar. 2003) 20–23, available at http://www.sieps.se/sites/default/files/3-20031.pdf (last visited Aug 15, 2015).

23 national strategic plans and reoriented towards three main objectives: achieving full employment; improving quality of work and labor productivity; and strengthening social and territorial cohesion. Furthermore, “quality of work” was replaced by the simpler concept of “flexicurity,” a term that is understood to mean a balance between the increased need of companies for flexibility and the need of workers for security in terms of employability and occupational advancement.112 This reassessment of the Lisbon Strategy seems to reveal a weak layering process. Policies were not redesigned from scratch but were simply refocused on the economic rationale of EMU. With the recent economic crisis, this biased approach has been seriously questioned.113 In particular, the inability of the Lisbon Strategy to handle socioeconomic shocks shed light on its ambiguous nature and experts have questioned whether weak normative instruments were the right means by which to promote policy learning. Even though the OMC produced learning across Europe, one may wonder whether the robust European social models will prove equally resistant to the greater pressures exerted by the economic crisis. 2. Reaffirmed Asymmetric Governance The goal of market integration did not require any harmonization of social and employment policy. As argued above, EU social policy was defined as a secondary priority of the EU construction, relegated to the subordinated realm of national decision-making.114 The Union only agreed on a framework of basic minimum standards intended to counterbalance the destructive and downward spiral stemming from EMU. The idea was to provide a minimal bulwark against using low social standards as an instrument of unfair competition. Against this backdrop, the EU has always had very little power to successfully formulate social or employment regulations. Even when it had the legislative competence, Member States could hardly reach a consensus. One way to overcome legislative deadlocks and influence national-level systems was to shift away from the classic legislative method towards a more coordinated model.115 This is how soft modes of learning emerged as the only response available to counter the pressure stemming from the EMU. The idea of coordinated learning began to take shape around the time of the Green Paper on Partnership for a New Organisation in 1994.116 The Commission suggested a “move from rigid and compulsory systems of regulations to more open and flexible legal frameworks.”117 A few years later, the Title on Employment institutionalized a new coordination model of governance away

112

G. Raveaud, The European Employment Strategy: Towards More and Better Jobs?, 45 JCMS J. COMMON MARK. STUD. 411–434 (2007). 113 European Commission, Economic Crisis in Europe: Causes, Consequences and Responses, (Directorate-General for Economic and Financial Affairs, European Economy No 7, Sept. 2019), available at http://ec.europa.eu/economy_finance/publications/publication15887_en.pdf (last visited Aug 15, 2015). 114 D. Ashiagbor, EMU and the Shift in the European Labour Law Agenda: From “Social Policy” to Employment Policy’, 7 EUR. LAW J. 311–330 (2001). 115 K. A. Armstrong, The Character of EU Law and Governance: From “Community Method” to New Modes of Governance, CURR. LEG. PROBL. 1–36 (2011). 116 European Commission, Green Paper: Partnership for a New Organization of Work, Apr., 1997. COM (97) 128, 5. 117 Id. para 44.

24 from “social law and legislative initiatives towards soft law or rather policies.”118 Ashiagbor has described this attempt to introduce greater flexibility to labor law as a move from employment law to employment policy, in which national labor law systems have become more flexible. One may therefore argue that this shift exemplifies a kind of institutional drifting whereby employment policy is weakened by monetary and fiscal policy.119 Accordingly, it was argued that Lisbon’s weak learning model was incapable of countering deregulatory pressures exerted by EMU. A common view was that Lisbon did not “have the means of its ambitions”120 and lacked “the real means of a proactive macro-structural policy mix (…) implementing structural reforms without macroeconomic governance.”121 It was assumed that Lisbon could not encourage social progress merely via soft learning while the ECB and the Commission exerted deregulatory pressures via fiscal and monetary hard law. Monetary stability and sound money were still seen as primary objectives, which had to be attained at any cost, even if it meant making labor and welfare more flexible. This is why many have considered Lisbon a failed attempt to rebalance EMU or as a hidden strategy to reaffirm the asymmetry between economic and social governance. 3. Evidence of Drifting? According to early evidence on the OMC, “New Governance” instruments had only limited value in promoting effective short-term implementation of a particular policy. There was only limited evidence of direct impact in the form of qualitative indicators endorsed by the Indicators sub-group of the EMCO,122 which concerned only a limited number of Member States.123 For Hemerijck and Visser, “learning” was neither a sufficient nor a necessary condition for policy change. Learning from other countries is but one possible factor among others in the change of social policy arrangements. It was said that “learning” does not necessarily improve performance, particularly if “learning” does not rely on one’s own experience.124 Theoretically, learning may be secured where stakeholders are brought together in deliberative problem-solving settings; where policy networks are enlarged; where decentralized experimentation is encouraged; where information is precise and available; and where actors are encouraged to compare their results with those of the best performers.125 118

Ashiagbor, supra note 114. Fitoussi & Eloi supra note 104 ; J. Creel, E. Laurent, and J. Le Cacheux, Delegation in Inconsistency: The “Lisbon Strategy” Record as an Institutional Failure, (Observatoire Français des conjonctures économiques (OFCE), Sciences Po 2005, Document de Travail No 2005-07, Jul. 2005) avialalbe at https://hal-sciencespo.archives-ouvertes.fr/hal-00972772 (last visited Aug 15, 2015). 120 Fitoussi & Eloi supra note 104, at 14. 121 J. Creel, E. Laurent, and J. Le Cacheux, supra note 119. 122 These indicators allow the tracking process of the employment guidelines as well as comparisons between countries performance. 123 See also M. Mailand, The Uneven Impact of the European Employment Strategy on Member States’ Employment Policies: A Comparative Analysis, 18 J. EUR. SOC. POLICY 353 –365 (2008); A. Schäfer, A New Form of Governance? Comparing the Open Method of Co-ordination to Multilateral Surveillance by the IMF and the OECD, 13 J. EUR. PUBLIC POLICY 70–88 (2006). 124 M Ferrera & A Hemerijck, Recalibrating Europe’s Welfare Regimes, in GOVERNING WORK AND WELFARE IN A NEW ECONOMY: EUROPEAN AND AMERICAN EXPERIMENTS 88–128 (Zeitlin & Trubek eds., 2003); A. Hemerijck & J. Visser, Policy Learning in European Welfare States, UNPUBL. PAP. LEIDEN UNIV. UNIV. AMST. 49 (2003). 125 M. Büchs, Directly-Deliberative Polyarchy: a Suitable Democracy Model for European Social Policy?, in THE ROLE OF INTERNATIONAL ORGANIZATIONS IN SOCIAL POLICY: IDEAS, ACTORS AND 119

25 These conditions are given in the OMC framework, but they remain speculative and data were too scarce to verify whether learning took place in practice.126 While some elements were easy to find (available information, stakeholders’ involvement), others remained rather vague or simply missing (deliberative problem-solving, decentralized experimentation). 127 However, while the launch of the EMU has rendered any return to traditional Keynesian social policy traditions unlikely, it is nevertheless possible to challenge the pessimistic predictions regarding the future of social and employment policy.128 Recent findings129 on the operation of the EES have demonstrated significant substantive and processual learning influence. Even though policy learning has not been easily observable, we have found concrete evidence resulting from OMC cycles. Most researchers agree that adjustments of domestic settings are not always apparent and that we have to look beneath the legal surface for a better understanding of the various learning scenarios. Most changes do not take the form of “legal transplants” and yet they are no less relevant than direct legislative changes. The type of influence the OMC exerts is discursive or cognitive, involving mostly single-loop cognitive shifts. On some occasions learning goes beyond the single-loop stage, assuming the more complex form of agenda (re)framing or direct policy shifts. III. EMU PATH-DEPENDENT LIMITATIONS IN CRISIS In this section we shall discuss the extent to which the institutional and legal path dependency and stickiness (A) of Maastricht helped to lay the groundwork for the euro crisis. (B) A. Accounting for EMU Institutional Stickiness While a gradual adaptation of the EMU institutional framework apparent, the euro area has not been moving neither towards a radically new governance framework nor towards a genuine rebalancing. Institutional innovations are still entrenched in the path-dependent Maastricht trajectory. In sum, the institutional framework did not adapt sufficiently to external economic conditions. The EMU has only engaged in a drifting of social institutions and a strengthening – displacement and redirection – of the monetary and fiscal dimension. This lack of profound institutional adjustment can be explained by two filtering IMPACT 49 (R. Ervik, N. Kildal, & E. Nilssen eds., 2009); S. Deakin, Reflexive Governance and European Company Law, 15 EUR. LAW J. 224–245 (2009). 126 J. S. Mosher & D. M. Trubek, Alternative Approaches to Governance in the EU: EU Social Policy and the European Employment Strategy, 41 JCMS J. COMMON MARK. STUD. 63–88 (2003), D. M. Trubek and J. S. Mosher, New Governance, EU Employment Policy, and the European Social Model, New York University, Jean Monnet Working Paper Working Paper (Sep. 2001). 127 S. Kröger, When Learning Hits Politics or: Social Policy Coordination Left to the Administrations and the NGOs?, 10 EUR. INTEGR. ONLINE PAP. EIOP 21 (2006); D. Natali, The Lisbon Strategy, Europe 2020 and the Crisis in Between, in EUROPE 2020: TOWARDS A MORE SOCIAL EU? (E. Marlier & D. Natali eds., 2010). 128 A. Verdun, An “Asymmetrical” Economic and Monetary Union in the EU: Perceptions of Monetary Authorities and Social Partners, 20 J. EUR. INTEGR. 59–81 (1996). 129 J. Zeitlin, Towards a Stronger OMC in a More Social Europe 2020: A New Governance Architecture for EU Policy Coordination, in EUROPE 2020: TOWARDS A MORE SOCIAL EU? 253–273 (E. Marlier, D. Natali, & R. Van Dam eds., 2010),); M. HEIDENREICH & J. ZEITLIN, CHANGING EUROPEAN EMPLOYMENT AND WELFARE REGIMES: THE INFLUENCE OF THE OPEN METHOD OF COORDINATION ON NATIONAL REFORMS (2009).

26 mechanisms: first, we refer to instrumental isomorphism which is the process of learning from policy failure (1); and secondly, it is the ability of Advocacy Coalitions to stale the process of institutional progression (2). 1. Instrumental Isomorphism Instrumental isomorphism entails both normative re-evaluation and cognitive shift in the sense that policy-makers draw lessons from experience and past mistakes. Most of the organizational literature emphasizes crisis and failure as tipping points of learning.130 In the learning literature it is also argued that learning is triggered by performance failure, which becomes opportunity for coalitions and policy-makers to push forward new ideas. They adopt institutional changes because they are convinced that such changes are the most appropriate solution to a given problem. 131 It is a form of rational policy-making, evidence-based process that can be characterized as an updating process based on “Bayesian updating.”132 Policy-makers act as goalorientated problem-solvers, consciously seeking to build consensus on institutional alternatives based on prior beliefs, expertise and experience. 133 This process of lesson drawing is highly reliant on objective data. Strong evidence is needed to persuade (skeptical) policy-makers to engage in change. That said, the relative stability and favorable economic conditions might explain why policy-makers only introduced small changes. The small number of on-path changes took place to address the economic and political interconnectedness between euro area economies, but they did not go far enough insofar as the EMU flaws were not exposed. Favorable economic conditions, notably the “gridlock economies: of moderate growth and price stability – the main objective of the EMU – played thus a significant role in hampering the EMU to address its flaws. 2. The Role of Advocacy Coalitions A slightly different strand of applied policy research associated with cognitive psychology literature considers that actors update their beliefs in accordance not with hard evidence, but with cognitive shortcuts.134 Indeed, policy-makers tend to draw disproportionate conclusions from limited empirical data to find what they want to find. In sum, policy-makers only engage in a highly path-dependent lesson-drawing process and often rely on the “lessons” learned from “success stories” to fix problems, thus reinforcing already-existing theoretical paradigms. It is more a process of layering than genuine institutional redirection.135 This was well exemplified during the negotiations of the Lisbon Treaty in 2008. 130

M. DIERKES ET AL., supra note 14, M. Easterby-Smith, M. Crossan & D. Nicolini, Organizational Learning: Debates Past, Present and Future, 37 J. MANAG. STUD. 783–796 (2000). 131 J. G. MARCH & J. P. OLSEN, REDISCOVERING INSTITUTIONS: THE ORGANIZATIONAL BASIS OF POLITICS (1989). 132 C. M. Radaelli, Measuring Policy Learning: Regulatory Impact Assessment in Europe, 16 J. EUR. PUBLIC POLICY 1145–1164 (2009). 133 C. J. Bennett & M. Howlett, The Lessons of Learning: Reconciling Theories of Policy Learning and Policy Change., 25 POLICY SCI. 275–294 (1992); H. C. Jenkins-Smith & P. A. Sabatier, Evaluating the Advocacy Coalition Framework, 14 J. PUBLIC POLICY 175–203 (1994); H. Heclo, Social Policy and Political Learning, in MODERN SOCIAL POLITICS IN BRITAIN AND SWEDEN (Heclo ed., 1974). 134 K. WEYLAND, BOUNDED RATIONALITY AND POLICY DIFFUSION: SOCIAL SECTOR REFORM IN LATIN AMERICA (2007). 135 S. Dahan, The Legal Framework for New Economic Governance and Its Implications for Wage Policy, 16 CAMB. YEARB. EUR. LEG. STUD. (2014).

27 First, the finance minister refused to discuss a genuine revamping of the EMU insofar as it was assumed that the ordoliberal/neoliberal model of Maastricht was a “success story.” Second, tensions aroused between Germany and France on the issue of finding a political counterweight to the ECB, something impossible in the Monetarist German conception of a currency union. That said, the institutionalist approach helps to illustrate the institutional resilience of the EMU and its insufficient capacity to engage in institutional displacement in order to adjust to economic conditions. Advocacy coalitions cannot opt for “breakdown and replacement” even in times of crisis.136 Therefore, a revamping of the EMU setup cannot be expected and any change will be rather gradual and dominated by the redirection of existing institutions. In particular, it has become clear that, at least in the near future, no new institutions will be created and the repartition of competence between the EU and Member States will not be affected. However, these small changes and institutional layering were not sufficient to address the EMU flaws that have become evident in the aftermath of the financial crisis. While euro-area economies have become more interconnected, the EMU structures did not provide with the right instruments to absorb economic shocks. In fact, this paper shares the views that the EMU institutional stickiness contributed to the development of the crisis. B. The Institutional Causes of Macroeconomic Imbalances In contrast with the deregulatory expectations discussed above, the euro fostered neither real convergence nor a race to the bottom in labor law standards. In fact, the opposite was observed: the EMU has exacerbated the pre-crisis diverging “growth models” that were revealed to be unsustainable across the eurozone. Specifically, the ECB uniform monetary policy (1) combined with distorted fiscal learning within SGP (2) and absence of wage coordination (3) has favored the development of unsustainable growth models resulting in wide macroeconomic imbalances. 1. Supranational Uniform Monetary Policy: Catalyst for Diverging Growth Paths As pointed out in Section 2, after the collapse of the Bretton Woods fixed exchange rate regime, monetary policy was declared neutral and high price stability policy was regarded as a precondition of economic growth. According to this view, central banks had to be independently responsible for price stability only, government for (de)regulation, and the unions for (low wages). In line with this thinking, the founders of the EMU assumed that the economy is always in a state of equilibrium and that monetary policy serves the economy best when it follows a low-inflation path. This led to the consensus in the Maastricht Treaty that the primary objective of an independent ECB is to maintain price stability (Article 119(2) TFEU). No mention was made of any other objectives of macroeconomic management such as high employment levels or financial stability.137 Moreover, the independence of the ECB has not remained confined to its relationship with political institutions. Rather, the ECB has repeatedly overstepped its mandate by advancing its own neoliberal agenda and thus assuming an ideologically independent stance.138 136

P. A. Hall & C. R. Taylor, Political Science and the Three New Institutionalisms, 44 POLIT. STUD. 936–957 (1996). 137 M. Sawyer, Alternative Economic Policies for the Economic and Monetary Union, 32 CONTRIB. POLIT. ECON. 11–27 (2013). 138 Id. 16.

28 This monetarist-asymmetric governance model finds its main inspiration in the Bundesbank anti-inflationary system.139 Due to inflation fears, the Bundesbank responded to output gaps asymmetrically in different economic situations. When output gaps were positive – when the economy grew faster than potential output – the Bundesbank feared inflationary pressure and reacted strongly by raising interest rates. By contrast, when the output gap was negative the Bundesbank did significantly reduce interest rates and did not counter recession. This is how Germany became a leader in price stability and the learning model in Europe. Several countries (including Austria and the Netherlands) pegged their currencies directly to the Deutschmark while others were influenced by the bank policy through the ERM Mechanism.140 This Bundesbank-style approach was then transferred to the ECB. The pursuit of other considerations became conditional on price stability. Once price stability had been achieved, Member States would learn from differences and national divergences would disappear. Initially the ECB single interest rate did fulfill the hopes of its supporters. National inflation rates, which had steeply declined in the run-up to the euro, continued to remain significantly lower that they had been in the 1990s (Figure 1). Furthermore, the financial markets honored the elimination of devaluation risks so that interest rates of government bonds and commercial credit declined steeply to the German level across the EMU (Figure 2). However, after this pre-1999 convergence phase EMU members embarked on differing economic growth paths. According to one explanation, the ECB could not reproduce the success of the Bundesbank because the union did not fulfill the main preconditions of an “optimum currency area” (OCA):141 Indeed, the EMU has little labor mobility and lacks fiscal transfer mechanisms.142 This argument did not have much influence on EMU design because it was assumed that there would be endogeneity in the fulfillment of the criteria.143 In other words, given the encouraging efforts of the pre-1999 convergence phase, and assuming the SGP would in effect work, it was expected that the increasing integration of capital, goods and markets would lead to the fulfillment of the OCA criteria, ensuring convergence of prices, wages and business cycles.144 One shortcoming of this decontextualized approach is that it overlooks the fact that pre-crisis imbalances between countries can be reinforcing rather than self-correcting under a fixed exchange rate system. The problem was that little attention was paid to these imbalances or national differences or how they would be affected in a single-currency context. This paper adopts a path-dependent approach and takes into account national differences in institutions, law, history, and policy outlook to evaluate the 139

Schettkat supra note 21 at 195. D. MARSH, THE BUNDESBANK: THE BANK THAT RULES EUROPE (1992). 141 Indeed, an optimum currency area is usually characterized by capital and labor mobility and the availability of fiscal transfer in case of asymmetric shock. See B. EICHENGREEN & P. H. LINDERT, THE INTERNATIONAL DEBT CRISIS IN HISTORICAL PERSPECTIVE (1992); R. A. Mundell, A Theory of Optimum Currency Areas, 51 AM. ECON. REV. 657–665 (1961). 142 K. Armingeon & L. Baccaro, Political Economy of the Sovereign Debt Crisis: The Limits of Internal Devaluation, 41 IND. LAW J. 254–275 (2012); W. H. Buiter, Optimal Currency Areas: Why Does the Exchange Rate Regime Matter?, (Centre for Economic Performance, London School of Economics and Political Science, Working Paper, Jul. 2000), available at http://eprints.lse.ac.uk/20178/ (last visited Jul 12, 2014). 143 R. E. BALDWIN & C. WYPLOSZ, MONETARY POLICY IN THE EURO AREA (2006). 144 Id. 140

29 convergence/learning influence produced by the euro. From this perspective, we argue that convergence expectations could not be achieved for two reasons. Firstly, the impressive learning results achieved by the “unlikely candidates” did not really address the structural and institutional differences that had originally caused economic divergence.145 As pointed out above, learning is a path-dependent process and therefore once the euro was introduced, these differences re-emerged.146 Secondly, the “one-size-fits-all” monetary policy reflected average conditions in the eurozone and could not be adapted to specific national conditions. Specifically, the transmission mechanism of monetary policy – how the single interest rate affects economic activity, wage and price inflation – was not reflected in the ECB interest rate.147 The ECB has not operated upon a path-dependent approach and has designed its interest-rate policy with the whole eurozone in mind, despite the diverging inflation rates. This seemed to be acceptable at the time of Maastricht, as the euro was expected to generate convergence and thereby reducing the degree to which the interest rates had differential effects.148 However, the “one-size-fits-all” approach had the opposite effect. Instead of fostering learning, the euro encouraged diverging growth paths and differential inflation rates. Indeed, the real interest rate – the nominal interest rate set by the ECB, minus country-specific inflation rates – became lower in countries with high inflation and higher in countries with low inflation. On the one hand, for countries with belowaverage economic growth and inflation rates, the interest rates were too high, consequently depressing weak economies even further. For instance, Germany was the main victim of this procyclical model. While nominal interest rates converged, inflation rates remained low (Figure 1). As a result, interest rates in Germany became the highest in the eurozone and growth remained lower than in almost all EMU member economies (Figure 3 and 4). On the other hand, for countries with aboveaverage inflation rates, the nominal interest rate policy was too accommodating because the ECB target was lower than the actual national inflation rate. The real interest rate became extremely low, even dropping into negative territory in the periphery, consequently feeding high economic growth (Figure 3). In sum, the onesize-fits-all monetary policy amplified already-existing divergences and contributed to the creation of two-speed growth models.149 Against this background, there have been significant divergences between countries in the driving force of growth. On the one hand, the sudden fall of nominal interest rate to German levels fed into credit-financed domestic demand in highinflation countries (the so-called periphery or debtors countries). This growth model was highly dependent on an expansion of private credit and on increasing asset prices in the market for commercial and residential investment.150 It has become the key 145

F. W. Scharpf, Monetary Union, Fiscal Crisis and the Preemption of Democracy, (Max Planck Institute for the Study of Societies, MPIfG Discussion Paper No 11/11, May. 2011) available at http://www.econstor.eu/handle/10419/48648 (last visited Jan 26, 2015). 146 C. Wihlborg, T. Willett & N. Zhang, The Euro Crisis: It Isn’t Just Fiscal and it Doesn’t Just Involve Greece, 11 WORLD ECON. (2010). 12/15/2015 8:02:00 PM147 As argued by Sawyer, “The general inflationary climate … depends on historical experience and expectations, the structure of the labour market, relationships between employers, employees and the State etc., and differences in inflation, unit labour costs and competitiveness are relevant for the evolution of currency union”. See Sawyer, supra note 137, at 16. 148 Palley, supra note 4; Deakin, supra 4, at 16. 149 H. Enderlein, Break it, Don’t Fix it!, 42 JCMS J. COMMON MARK. STUD. 1039–1046 (2004). 150 To a lesser extent in Portugal. See A. Afonso & R. M. Sousa, Assessing Long-Term Fiscal

30 source of demand growth in Britain, Ireland and some continental European countries such as Spain, Greece and parts of Eastern Europe where household debt has increased dramatically in the last decade (Table 1). These countries have provided the main source of growth (and wage) inflation since the introduction of the euro and have typically run substantial account deficits (see paragraph 2). On the other hand, low-inflation countries (creditors or core economies) have followed policies of export-led growth that are dependent on targeted investment in capital goods, public support for training and labor force upgrading, and wage moderation. These countries did not experience a significant rise in household incomes, thanks to wage restraint and lower growth (Table 1). However, their competitiveness did improve, whereas the GIIPS, unable to counter wage inflationary pressure, lost competitiveness at the same time. (We shall return to these issues in the following section). These developments form part of a more global pattern that became apparent with the re-emergence of neoliberalism in the 1980s. In this model, export-led countries have been relying on the maintenance of demand in finance-led countries, a dependence that was further fostered by capital outflows in the export-led countries.151 In this paper, we argue that this dynamic has been intensified by the legal framework of the EMU. As we have described, the ECB uniform monetary policy has played a significant role in the development and amplification of this unbalanced model of growth. However, the ECB problem constitutes only one piece of the puzzle. In the following section, we argue that this damaging model was further amplified by two other features of the EMU framework: firstly, the SGP failed to detect and correct these diverging paths, and secondly, this procyclical model was heightened by the absence of wage determination. 2. Distorted Fiscal Policy: Lack of Real Convergence and Deterioration of Public Finance The design flaws of the EMU were also present in the distorted model of Maastricht. In particular, the convergence criteria are said to be distorted and obscured by the Union’s obsession with price stability. As argued in Section II, the EMU neoliberal project changed the fiscal-monetary balance in Europe to diminish the roles of the government and enhance the power of the market. Previous systems ensured fiscal dominance whereby central banks served the government. In the new system, national governments are prevented from incurring debts that could impose a “moral hazard” problem for other Member States. Moreover, governments are prohibited from accessing either their own central banks or the ECB to finance their budgets (Article 123 TFEU) and from receiving any kind of financial aid from the ECB or Member States (Article 125 TFEU, the so-called “no-bailout clause”).152 In this context, monetary stability was perceived as the sine qua non of credibility where the euro and sovereign borrowing were concerned. Since 2002, therefore, when the single currency reached its objective of stabilizing inflation at a low level around 2% annually, the financial community granted the same interest rates for all the public debts in the eurozone. Even the periphery – Greece, Portugal and Spain – which had to pay very high interest rates in the 1990s were granted low interest rates. However, we believe this convergence of interest rates to be actually a function of

Developments: Evidence From Portugal, 18 APPL. ECON. LETT. 1–5 (2010). 151 S. Deakin, “Euro Crisis and Labour Law”, February 5, 2013; Palley, supra note 4. 152 Palley, supra note 4; Deakin, supra note 4.

31 several misleading expectations associated with Maastricht.153 Firstly, it was assumed that the euro would bring a quasi-convergence of inflation rates across the eurozone. Secondly, it was expected that all euro members could maintain their competitiveness without currency devaluation. Finally, the SGP and Maastricht were to be enforced by market discipline. Indeed, financial markets were supposed to take account of the European treaties, particularly the prohibition of any fiscal or financial solidarity between member countries of the eurozone. Achieving convergence proved, however, to be problematic in practice. The SGP only achieved formal convergence without fostering real convergence. Indeed, the SGP has focused mainly on deficit and debt figures that resulted in a high degree of apparent convergence. In effect, at the onset of the crisis all eurozone members besides Greece were in compliance with the Maastricht criteria, and several Member States which had incurred soaring budget deficits after 2007, including Ireland, Portugal and Spain, had surpluses as well as national debt levels at historic lows (Table 2 & 3). However, beneath this formal compliance with Maastricht there was a lack of real convergence.154 Most convergence expectations predicted by the Maastricht founders turned out to be inexact, and neither Maastricht nor the markets were able to foresee or correct diverging growth paths. As argued above, the core countries have exported to the periphery while the south has relied mainly on finance-led growth. As a result, current account deficits, and specifically the balance of trade, widened in the periphery as domestic production systems could not match the boom in domestic demand, whereas the core economies generated a growing account surplus (trade surplus).155 This is a perfect example of macroeconomic imbalance in which current account deficits had to be funded through capital inflows, and current account surpluses involved capital outflows. In other words, export-led strategies leading to current account surpluses in the core economies had to be matched by current account deficits in finance-led countries. Another problem that we discuss below is that current account imbalances are usually symptoms of a significant deterioration in cost competitiveness in the export sectors, which is usually the result of excessive reliance on capital inflows. We believe that the absence of real convergence is partly the result of the SGP price stability bias. The convergence criteria in effect paid no attention to non-fiscal real variables such as the validity of the exchange rate at which countries accessed the eurozone; prevailing current account deficits or surpluses; or differences in inflation rates across the eurozone. Similarly, Maastricht included no provisions regarding convergence of business cycles and economic conditions, nor any that addressed the implications of the ECB monetary policy on imbalances and inflation rates. We do not argue that the SGP is responsible for the widening of imbalances. The ECB monetary policy and the removal of the exchange rate have in fact played a primary role in these developments by allowing the periphery to borrow from surplus States at much lower interest rates than previously. However, the SGP played an important role in the crisis in that neither the formal criteria nor the financial markets foresaw or corrected national vulnerabilities. R. Boyer, “Origins and Ways Out of the Euro Crisis: Supranational Institution Building in the Era of Global Finance” (2013) 32 Contributions to Political Economy 97–126. 154 Deakin supra note 151; Palley, supra note 4. 155 The current account position is composed of the trade position and the net income flows, and the latter interest payment on borrowing. The largest component of the current account calculation is generally the balance of trade. 153

32 As long as the world economy was growing at a high rate, these imbalances went unnoticed. The reversal occurred in the spring of 2007 in the derivative markets of a small segment of the US mortgage market, the sub-prime market. The collapse of a major financial institution (Lehman Brothers) triggered a mortgage crisis that spiraled into a catastrophic financial crisis. Private financial markets froze and several financial institutions suffered liquidity problems. As a result, governments had to intervene to stabilize the private market and to recapitalize several banks. Eurozone governments also had to take significant fiscal measures in order to contain the recession. Impressive recovery plans supplemented the automatic stabilizers to prevent the collapse of economic activity. It was only in late 2009 and early 2010, when the fiscal cost of the financial crisis became obvious, that the financial markets began to worry about public debt in Europe. Ultimately, the event that triggered the euro crisis turned out to be the implosion of the Greek economy. Hungary, Romania and the Baltics had already turned to the IMF but the Greek problem merited special attention because its policy options were limited by their eurozone membership. The situation in Greece was like Europe’s Lehman catastrophe, which triggered the so-called sovereign debt crisis. The financiers considered the Greek problem alarming and quickly readjusted their criteria for assessing the financial health of the eurozone members. They realized that Greece (and Portugal) had accumulated deficits above the limit permitted by Maastricht since their accession to the eurozone and generalized the problem to the rest of the eurozone. They did this even though fiscal profligacy was not a problem in Spain and Ireland, which had demonstrated a high degree of fiscal responsibility in the six years before the crisis (Tables 2 & 3). Both Spain and Ireland ran budget surpluses for much of the five-year period (2000-2007) and their average deficits were smaller than those of Germany from 2002 to 2005. The “fiscal profligacy” problem only developed as a function of the public bailout of the banking systems that had resulted from the financial shock of 2008. It is thus inaccurate to describe the eurozone crisis as one of fiscal irresponsibility. Rather, it is the asymmetric model of the EMU that failed catastrophically, plunging the periphery into a budget deficit. The public debt crisis is only the final stage of the process. The real issue was that most countries in the periphery found themselves in a vulnerable position defined by current account deficits and extreme dependence on capital inflows at the onset of the crisis. As discussed above, these vulnerabilities were the result of the asymmetric management of the EMU. Specifically, the ECB single interest rate fuelled credit-led growth in the periphery while the SGP failed to foresee and correct the development of imbalances. The problem was not fiscal but structural. The financial crisis evolved into a sovereign debt crisis because of a publicprivate debt transmission process that was assumed by governments. The increase in budget is only the direct consequence of banks stabilization, counter-cyclical fiscal stimuli, increase in social protection spending due to the rise in unemployment, and the loss of tax revenues during the crisis. Tables 7 and 8 show the budget deficit/GDP and ratios of public debt to GDP during 2006-2010. In all countries there is a significant increase after the crisis, but most notably in both Spain and Ireland as well as in Germany where a budget surplus of 2007 turned into a deficit. As for debt/GDP ratios, most countries had very modest figures prior to 2007. 3. Weak Labor Policy: Exacerbating Macroeconomic Imbalances The diverging growth paths described above are not only the result of ECB

33 inflationary pressures and distorted fiscal convergence. Labor policy (un)coordination and particularly wage arrangements have also contributed to the widening of divergences between the core and the periphery. On the one hand, high finance-led growth in the periphery was accompanied by fuelling (wage) inflation, thanks to the ECB interest rate but also to the ineffectiveness of policy learning mechanisms in the field of wage policy. Wages grew faster than productivity and national wage-setting arrangements had little leverage to mitigate the inflationary pressures coming from the ECB.156 On the other hand, the unnecessarily restrictive monetary policy imposed on low-inflation economies was accompanied by strong wage-moderation mechanisms, thereby worsening the deflationary pressures on these economies.157 This pro-cyclical dynamic eventually catapulted the competitiveness of Germany and the periphery onto diverging paths. This section will show that the absence of a common learning framework on wage determination played a crucial role in this pro-cyclical process. More precisely, while the core economies – Austria, Belgium, Finland, France and Germany – were able to impose wage restraint through wage-setting mechanisms, the periphery – Ireland, Italy, Portugal and Spain – lacked the legal and institutional capacity to restrain wage growth and consequently lost competitiveness relative to the core. The following sections are organized as follows: The section (a) claims that the EMU asymmetric framework featured a limited wage policy model which proved unable to counter inflationary pressures. Sections (b) and (c) argue that macroeconomic imbalances were amplified by two different wage coordination paths (Figure 9), with above-productivity increases in the periphery and below-productivity increases in the core. Section (d) argues that these wage (and price competitiveness) developments find their origin in the absence of wage-policy mechanisms at the EU level, which resulted in diverging wage-setting institutions. a. The EMU Crisis and Wage Policy Learning This section analyses the eurozone competitiveness crisis in the context of monetary integration and domestic wage-setting institutions. Before the introduction of the euro, the political economy of the prospective eurozone candidates was a robust wage-restraint system closely pegged to the German model. Aggregate nominal wage cycles of most candidates were closely calibrated to German wages through the interaction of wage-setters and central banks. National central banks usually responded to (wages and prices) inflation by threatening to raise the interest rates. Wage restraints usually targeted the sheltered sector, particularly wages in the highly unionized public sector, and forced it to follow wage rates adopted in the exposed sectors, where external competition imposed a strong deflationary pressure. Predictably, many of the prospective candidates faced a period of intense social conflict. This disciplinary process translated into a tightly organized wage system in which exposed sector wages were synchronized to the German wage model and the sheltered sector hierarchically synchronized to the exposed wage sector.158 However, once the euro was introduced the disciplinary task of the NCBs was not transferred to the ECB. Maastricht only transferred monetary policy to the ECB 156

Deakin supra note 151; B. Hancké, Worlds Apart? Labour Unions, Wages and Monetary Integration in Continental Europe, Institute for Advanced Studies, IHS Political Science Series No 128 (Feb. 2012). 157 Id.; Scharpf, supra note 145. 158 Hancké, supra note 156.

34 without a parallel centralization of wage-setting and fiscal policy. As we saw above, this new legal design gave rise to a procyclical macroeconomic management model with knock effects on wage growth. Indeed, the uniform interest rate has fed into asset price and wage inflation in the periphery while it has depressed wages and growth in the core of the eurozone. Furthermore, the EMU offered little legal means at the EU level to mitigate this effect. Firstly, these perverse effects could not be offset via the nominal exchange rate. Secondly, fiscal policy was hardly an option: as noted above, the SGP bias exacerbated the procyclical dynamic produced by the ECB by rewarding countries that had a surplus and punishing countries that had a deficit. Thirdly, the mismatch between ECB interest rates and domestic conditions posed serious challenges for wage-coordination insofar as the ECB could not retaliate against domestic unions to restrain excessive wage rates and competitiveness loss. Against this background, many observers predicted that a world in which a (European) central bank lacks credibility to coordinate wages would result in massive inflationary pressure.159 Other commentators went further, arguing that inflationaverse countries might opt for nominal wage flexibility and introduce labor market reforms that would lead to overall wage moderation or a race to the bottom in labor standards.160 The EMU experience demonstrates that neither of these scenarios was realized. Firstly, wage explosion did not take place. Wage aggregate growth remained moderate and there were very few signs of the massive inflationary pressures predicted. Secondly, despite pressures stemming from EMU, labor law systems seem to have remained relatively stable. In fact, there is strong evidence of progression, thanks to both national welfare traditions and the EU social impetus of the 2000s (Figure 7). As pointed out above, the first ten years of EMU gave rise to a rather unexpected “social moment” that began with the Jacques Delors’ presidency (19892008). This moment arrived with the Social Charter of 1989, the Amsterdam Treaty and the EES, which was defined as the “constitutionalization” of EU employment law.161 The next step in this direction was to be the Lisbon Strategy, which marked the institutionalization of flexible methods of coordination and learning. However, while we recognize the progress achieved by Lisbon, we argue that it has contributed to the widening of wage imbalances insofar as it was too weak to withstand the pressures exerted by the uniform ECB policy. 162 Specifically, the OMC has neither strengthened wage bargaining nor prevented union decline and we believe this might have contributed to diverging wage growth paths and competitiveness performance (Table 4 and Figure 8). Indeed, evidence shows that while most countries have witnessed some union decline since the 1970s, the core economies that had strong wage-bargaining institutions (particularly Germany, Austria, Belgium and France) implemented wage-restraint policies. At the same time, the periphery witnessed a weakening of their collective bargaining settings, which had the opposite effect, with wages rising above productivity level (Figures 8 and 9).163 Had the OMC D. Soskice and T. Iversen, “Multiple Wage-Bargaining Systems in the Single European Currency Area” (1998) 14 Oxford Review of Economic Policy 110–24. 160 See section 2. 161 C. Barnard, “The United Kingdom, the “Social Chapter” and the Amsterdam Treaty” (1997) 26 Industrial Law Journal 275–282. 162 See Section II on deregulatory pressures exerted on labour and social policy. 163 It is worth noting that union decline and wage bargaining weakening usually produce a different result, leading to a fall in the share of wages in national income. More of national income is 159

35 been able to steer wage bargaining mechanisms and to mitigate union decline we might have witnessed less divergence in wage growth and cost competitiveness. However, this was not a viable role for the OMC as Lisbon has always been torn between the primary economic logic of Maastricht and the secondary social component of Lisbon. As argued above, monetary policy was an exclusive Union competence vested in the ECB; economic policy was a shared competence; and social policy remained mostly un-harmonized even in areas that had implications for economic policy, such as wage determination. It was assumed that learning could be combined with economic deregulation and lack of genuine social and economic coordination. However, the fiscal and monetary bias of the EMU has rendered the OMC unable to prevent harmful side effects such as private indebtedness and asymmetric wage growth.164 In that sense, instead of addressing the Maastricht social deficit, Lisbon has helped monetary logic to destabilize the European Social Model.165 b. Export-led Countries Competitiveness Strategy (Germany) While the core economies suffered their worst slump since the 1970s, the effects of the crisis on their labor markets were less severe than they were in the periphery. In particular, the “German employment miracle” was held up for admiration all around the world. This “success story” is the result of a decade and a half of neoliberal political and economic transformation166 that was triggered by the monetary unification (along with other factors such as the sudden incorporation of East Germany). Since Germany is the largest economy in the eurozone, this section focuses on the German competitiveness strategy and addresses the questions of what exactly lies behind this success story and what role Germany played in the development of the crisis conditions. The role of Germany in the euro crisis is best understood in the context of the EMU legal framework. Before EMU, Germany’s economic domination was uncontested as a learning model for many other countries.167 Both its nominal interest rates and its real interest rates were at an all-time low. In 1999, however, Germany became the first victim of the ECB uniform monetary policy, and its learning influence and comparative advantage were lost as a result of the perverse procyclical effects of Maastricht.168 When the EMU nominal interest rates converged, Germany’s real interest rates became the highest in the eurozone. German economic growth was the weakest in the entire eurozone (Figure 4), and unemployment and social spending increased dramatically (Figure 10) while tax receipts decreased significantly. The EMU instruments of macroeconomic adjustment offered little room to going to profit and less to wages since wages are rising below productivity level. 164 C. Degryse, Social Developments in the European Union 2009 (ETUI and OSE 2010). 165 D. Natali, “The Lisbon Strategy, Europe 2020 and the Crisis in Between” in E. Marlier and D. Natali (eds), Europe 2020: Towards a More Social EU? (PIE-Peter Lang 2010). 166 G. Bosch, S. Lehndorff, and J. Rubery, European Employment Models in Flux: A Comparison of Institutional Change in Nine European Countries (Palgrave Macmillan 2009). 167 A. Johnston and B. Hancké, “Wage Inflation and Labour Unions in EMU” (2009) 16 Journal of European Public Policy 601–622. 168 Since the ECB’s 2 % aggregate inflation target only addresses variation in inflation across member countries, increasing inflation in some countries above the target entails will have to deflate to compensate to reduce the aggregate inflation rate to 2 %. As a result, Germany is forced to deflate if inflation rises in others.

36 maneuver in its response to this recession. Germany was legally unable to adjust via monetary or fiscal reflation, which had been acceptable options before Maastricht. On the monetary policy side, the Bundesbank could not lower interest rates to boost the economy. As for fiscal policy, Germany breached the SGP 3% threshold in 2003 by allowing an “automatic stabilizer” to operate, but that was not sufficient to absorb the economic shock. As a result, Germany opted for a third solution: a high degree of specialization of the industry169 combined with wage restraint in the leading industrial sectors. This supply-side strategy, which was aimed at protecting existing jobs, was made possible through several legal and structural transformations within the German labor market.170 The first reform was the dismantling of the German bargaining patterns that were intended to prevent the leading unions from extricating wage increases from 2000 onwards.171 Evidence shows that in the metalworking and chemical industries – the two leading sectors that are widely exposed to international competition – collective agreements imposed caps on wage increases. Likewise, in sheltered sectors that are not exposed to international competition – the construction industry, retail trade and the public sector, for instance – collective agreements also imposed strict pressure on wage levels (Figure 9 and 11). A second transformation concerns the Union density and coverage: both fell sharply in the 1990s and before the crisis (Table 4 and Figure 8).172. As a result, effective pay rises from 2000 to 2008 were on average 50% below the collectively agreed rates of increase, meaning that the nominal compensation per employee and ULC had fallen since the introduction of the euro (Figure 11, 13 and Table 6). Thirdly, the Red-Green government pushed towards supply-side policies between 2000 and 2005 with tax reforms applied to company profits and capital incomes, and on privatization of services previously supplied by the public sector. Fourthly, the deregulation of temporary and part-time employment (Figure 12) facilitated the expansion of atypical employment, the so-called “mini-jobs” that pay a maximum of EUR400 a month. 173 Finally, wage moderation might have been encouraged by the social security and fiscal reforms carried in 2003, the so-called “Hartz reforms.” Most importantly, unemployment insurance payments were reduced to 12 months, which compelled people to accept job offers with low skill requirements and low wages. While most economists agree that Germany’s export-led recovery after 2005 was due to wage repression and not Hartz reforms 174, we contend that these may have reinforced the trend by exerting downward pressure on many 169 G. Bosch, S. Lehndorff, and J. Rubery, European Employment Models in Flux: A Comparison of Institutional Change in Nine European Countries (Palgrave Macmillan 2009). 170 S. Lehndorff, “Before the Crisis, in the Crisis, and Beyond: the Upheaval of Collective Bargaining in Germany” (2011) 17 Transfer: European Review of Labour and Research 341–354. 171 Ibid. 172 J. Visser, “ICTWSS: Database on Institutional Characteristics of Trade Unions, Wage Setting, State Intervention and Social Pacts in 34 countries between1960 and 2007” (Amsterdam Institute for Advanced Labour Studies (AIAS) April 2013). 173 For more detail see K. Stone, “The Decline in the Standard Employment Contract: Evidence from Ten Advanced Industrial Countries” in H. Arthurs and K. Stone (eds), After the Standard Contract of Employment: Innovations in Regulatory Design (Russell Sage Foundation 2013); M. Dieckhoff, V. Gash, A. Mertens, and L. Romeu-Gordo, “Female Atypical Employment in the Service Occupations: a Comparative Study of Time Trends in Germany and the UK” in W. Eichorst and P. Marx (eds), Non-Standard Employment in a Comparative Perspective (Forthcoming 2013). 174 For more detail on German reforms and the Crisis, see M. Caliendo and J. Hogenacker, “The German Labor Market after the Great Recession: Successful Reforms and Future Challenges” (2012) 1 IZA Journal of European Labor Studies 1–24.

37 sectors of the labor market.175 As a consequence of these labor market transformations, the number of employees earning less than two-thirds of median pay increased by half since the mid-1990s, accounting for 22% of the working population. Mass protests against the welfare reforms and the defeat of the Red-Green government resulted in the introduction of minimum wage levels in a few sectors although statutory minimum wage is still not required by law in Germany.176 From a short-term perspective, this export-led model seemed to be a suitable response to the recession. Export demand and employment in the export industries and in the low-wage sector increased, thanks to the “impetus from the improvement in price competitiveness” resulting from wage recession177 (but thanks also to the high degree of specialization and product quality of the industry).178 More specifically, between 2001 and 2008 three-quarters of Germany’s growth was attributable to the export surplus, while domestic demand contributed one quarter.179 In the longer term, however, this has had several interconnected effects that have been damaging to the currency union as a whole. Firstly, competitiveness strategies of this kind produce current account surpluses that must be matched by current account deficits elsewhere, that is to say macroeconomic imbalances. Secondly, this aggressive neomercantilist strategy, in addition being very detrimental to the exports of the peripheral countries in crisis, has the effect of beggar-thy-neighbor in the field of wage and social standards.180 Indeed, the German success may influence other Member States to use a similar wage restraint approach as a means of adjustment to fill the competitiveness gap. This would likely result in a “race to the bottom” of the entire system. Thirdly, weak wage growth has prevented both a transfer of exportbased stimulus to the domestic market and a concomitant rise in imports which would have benefited other countries. In sum, the German response to the crisis of the 2000s contributed to the widening of current account imbalances between eurozone economies (Figure 5), but also to the weakness of the domestic market characterized by increasing inequality in the redistribution of income and capital. As shown in Figure 4 and Table 1, GDP growth rate and household disposable income has remained below the EU average since 2004. Against this background, one may question why this vulnerable model is so admired at home and abroad and held up as a model worth of emulation. 175 M. Erlinghagen and M. Knuth, “Unemployment as an Institutional Construct? Structural Differences in Non-Employment between Selected European Countries and the United States” (2010) 39 Journal of Social Policy 71–94; G. Bosch and T. Kalina, “Low-Wage Work in Germany: an Overview” in G. Bosch and C. Veinkopf (eds), Low-Wage Work in Germany (Russell Sage Foundation 2008). 176 G. Bosch and T. Kalina, “Low-Wage Work in Germany: an Overview” in G. Bosch and C. Veinkopf (eds), Low-Wage Work in Germany (Russell Sage Foundation 2008). 177 Deutsche Bundesbank, “Zur Entwicklung der Ausfuhr in den vier großen EWUMitgliedstaaten seit Beginn der Wëhrungsunion”, July 2011, Monthly Report. 178 G. Bosch, S. Lehndorff, and J. Rubery, European Employment Models in Flux: A Comparison of Institutional Change in Nine European Countries (Palgrave Macmillan 2009). 179 S. Lehndorff, “German Capitalism and the European Crisis: Part of the Solution or Part of the Problem?” in S. Lehndorff (ed), A Triumph of Failed Ideas: European Models of Capitalism in the Crisis (ETUI 2012) 81. 180 Scharpf, supra note 145, at 11; S. Lehndorff, “German Capitalism and the European Crisis: Part of the Solution or Part of the Problem?” in S. Lehndorff (ed), A Triumph of Failed Ideas: European Models of Capitalism in the Crisis (ETUI 2012) 81.

38

c. Deterioration of Competitiveness and Current Account Deficits in the Periphery For orthodox economists the crisis is only the consequence of a market disequilibrium problem and fiscal profligacy in the periphery. However, we believe that there were visible structural frailties before 2007 and that public spending was only a serious problem in Greece (and to a lesser extent in Portugal). 181 Indeed, the Greek model was mainly founded on the cheap-credit growth resulting from EMU accession. The main drivers of the GDP growth were rising domestic demand, which was based mainly on consumption, fuelled by rising real wages, rents and profits, and sustained public spending. Fiscal policy was indeed strongly expansionary in the postEMU period, exceeding the 3% limit on public deficit on several occasions. In 2007, Greece was in a catastrophic fiscal position with a public deficit exceeding 6% of GDP and a public debt at 107% of GDP. Most peripheral States were not in a similarly troublesome fiscal position on the eve of the crisis. However, they all concealed similar domestic vulnerabilities – current account deficits and competitiveness deterioration – that were exacerbated by their finance-led growth model, a lack of legal instruments at the EU level and weak wage-setting arrangements at the national level. Firstly, we discuss the role of finance-led growth in the deterioration of competitiveness positions and current account deficits. As argued above, the sudden availability of cheap finance after accession to the EMU combined with near-zero or even negative real interest rates182 fostered fragile finance-led growth in the periphery (although to a lesser extent in Portugal and Italy).183 In Spain and Ireland in particular, cheap credit fed into real estate investment leading to rapidly rising housing prices – a classic property bubble. As a result economic growth and employment rates increased sharply. Spain and especially Ireland were the new learning models of Europe until the financial shock of 2007.184 However, in spite of these positive results, high-growth models were already showing signs of fragility even before 2007. As shown above, the competitiveness positions of the peripheral States was deteriorating mainly as a result of increases in real wages (Figure 11) and ULC (Figure 13). Consequently, all those states (including Italy)185 shared a common symptom: significant current account deficits (Figures 11 & 5) caused by competitiveness deterioration (excessive wage growth) and rising imports. 186 Secondly, the EU had limited legal means by which to correct the widening competitiveness gap between the core and the periphery and by the same token the growing current account imbalances. The usual solution for a competitiveness crisis is exchange rate devaluation, but this is not possible in the EMU. Devaluation would have raised the price of imports and restored the competitiveness of exports. In M. Karamessini, “Sovereign Debt Crisis: An Opportunity to Complete the Neoliberal Project and Dismantle the Greek Employment Model” in S. Lehndorff (ed), A Triumph of Failed Ideas: European Models of Capitalism in the Crisis (ETUI 2012). 182 As shown above the ECB nominal interest rate resulted in too low (or even negative) real interest rates in high-inflation peripheral States. 183 This was true also but to a lesser extent in Portugal. 184 Nobody ever claimed that Greece and Portugal were learning models, but Ireland and Spain were hardly success stories either. 185 Italy also has serious competitiveness problems but did not experience GIPS finance-led growth. 186 The most common indicator is relative unit labour costs (ULC). 181

39 addition, exchange-rate devaluation would probably have been more effective than nominal wage cuts in reducing real wages, as wages tend to be sticky.187 Finally, devaluation would likely have increased domestic inflation and ultimately reduced the level of debt.188 Having ruled out exchange rate flexibility, eurozone members wishing to make adjustments found themselves with very little room to maneuver. Spain and Ireland attempted to slow wage inflation via fiscal restraint by running budget surpluses, but this proved insufficient (Figure 6). As we have shown, the SGP did not foresee the development of macroeconomic imbalances. Another option would have been to slow unsustainable finance-led growth of the periphery through monetary restraint. Had the ECB been able to set differentiated interest rates geared to national conditions, this might have helped to slow wage inflation. However, this solution can only work if growth is the only driver of wage inflation. Data show that growth can be an important driver of wage inflation but does not necessarily lead to wage inflation. High-growth countries such as Finland, for instance, did not experience wage inflation in the post-EMU period. As a result, monetary restraint would have been most effective in high-growth countries such as Ireland and Spain, where GDP growth was the main driver of wage inflation, but less effective in lowergrowth countries such as Italy and Portugal, where wage expansion was not the result of finance-led growth. In short, the phenomenon of wage inflation cannot be explained solely by inadequate monetary fiscal policy.189 This brings us to the third common problem in the periphery, the weakness of wage-coordination settings. Even though the EU has blamed labor regulations and union-wage militancy for this competitiveness problem, research shows that strong wage-bargaining institutions played a positive role in the development of the competitiveness problem. Evidence does not point to excessively strong labor law as the primary explanation for competitiveness loss. On the contrary, the disappointing wage performance of the periphery seems to be the result of the paucity of their wagebargaining institutions, whereas robust wage-bargaining institutions helped the core economies to enhance their competitiveness positions.190 We therefore argue that, as discussed below, diverging labor laws and wage-coordination institutions partly explain the phenomena of wage inflation and macroeconomic imbalances. d. The Role of National Wage-Setting Arrangements in the Deterioration of Competitiveness Positions The uniform ECB policy and the distorted SGP do not fully account for the problem of diverging (wage) growth paths and macroeconomic imbalances. We argue that the problem has been exacerbated by the absence of wage-policy learning instruments such as the OMC.191 Indeed, the introduction of the euro reopened the disparity within wage-setting models and might consequently explain why labor costs diverged between the periphery and the core bloc (Figure13). According to Hancké and Johnston, wage development dynamics can best be understood in light of the dichotomy between exposed and sheltered sectors since they K. Armingeon and L. Baccaro, “Political Economy of the Sovereign Debt Crisis: The Limits of Internal Devaluation” (2012) 41 Industrial Law Journal 254–275. 188 Ibid. 189 A. Johnston and B. Hancké, “Wage Inflation and Labour Unions in EMU” (2009) 16 Journal of European Public Policy 601–622. 190 Ibid. 191 Ibid. 187

40 usually exhibit different wage development patterns.192 On the one hand, in spite of the absence of the monetary threat, wage explosion does not usually occur in the exposed sector because of competitiveness concerns. On the other hand, the sheltered sector does not face the same pressures and is thus particularly vulnerable to wage inflation. Inflationary pressures in the sheltered sector are usually compensated for in the exposed sector as long as the productivity of the latter is high enough and wages grow at a moderate rate.193 However, in some cases the exposed sector has a reduced capacity to compensate either because it is too small relative to the sheltered sector or because the exposed sector also increases wages to above-productivity levels. This is what happened in finance-led countries where the constraints imposed by strong national central banks before the EMU were not replaced by hard legal incentives that linked wage-setting to productivity levels in both the sheltered and exposed sectors. On the one hand, the core economies (Austria, Germany, Belgium and France) have contained both the sheltered and exposed sectors’ wages by productivity increase through a tight coordination framework. Coordination instruments can take both legal and non-legal form. As indicated in Table 6, not all wage-restraint countries have binding institutional frameworks that constrain wage-coordination. For instance, in Germany and Austria the leading export (metalwork) sector unions exert pressure on the entire economy to synchronize sheltered and exposed wage-setting mechanisms. In Belgium, however, the 1996 law on wage competitiveness imposed a legal wage constraint that had to be respected by all sectors (Table 6). 194 This law set a ceiling on all wage increases in Belgium, mandating that no annual increase should raise the average wage above that of Belgium’s trading partners (France, Germany and the Netherlands).195 For that matter, France offers a particular tri-partite setting with an inter-sectoral wage coordination process relying on large multinational firms and the government. Unions are weak (low density) but the coverage is important and the sheltered (civil servant) unions are powerful and push constantly for aboveproductivity wage increases, often leading to social conflict.196 In the second group (Ireland, Spain, the Netherlands and Portugal), wage coordination channels were rather weak and institutional and legal constraints were absent. For Hancké and Johnston this explains why wages in both sheltered and exposed sectors, particularly in Italy and Portugal, diverged rapidly and why in some cases wages in the exposed sectors increased to above-productivity levels (Table 6 and Figures 9, 11, 13). This is exemplified by Ireland’s weak wage-determination process, which probably contributed to the development of the Irish wage-bubble.197 Before and after accession to the EMU, the government introduced time-irregular social pacts (the National Recovery plan in 1986 and a series of tripartite agreements agreed wage rates across the economy), but these agreements only minimally 192

Ibid. B. Hancké, “The Missing Link: Labour Unions, Central Banks and Monetary Integration in Europe” (2013) 19 Transfer: European Review of Labour and Research 89–101. 194 D. Neumark and W. Wascher, “Minimum Wages, Labor Market Institutions, and Youth Employment: A Cross-National Analysis” (2004) 57 Industrial and Labor Relations Review 4. 195 P. Pochet, “Belgium: Monetary Integration and Precarious Federalism” in M. Andrew and G. Ross (eds), Euros and Europeans (Cambridge University Press 2004). 196 B. Hancké and D. Soskice, “Wage-Setting and Inflation Targets in EMU” (2003) 19 Oxford Review of Economic Policy 149–160. 197 J. Wickham, “After the Party’s over: the Irish Employment Model and the Paradoxes of Nonlearning” in S. Lehndorff (ed), A Triumph of Failed Ideas: European Models of Capitalism in the Crisis (ETUI 2012). 193

41 constrained wages in the sheltered sectors. The main problem in Ireland and Spain was that the exposed sectors were too small to compensate for wage increases in the sheltered sectors, particularly real estate. Furthermore, when the real estate bubble burst the government was unable to compensate for the loss in the sheltered sectors, which consequently developed into a massive public debt in addition to a serious competitiveness problem. In the same vein, the Netherlands attempted to prevent the sheltered sector from creating inflationary pressures through temporary wage freezes in social pacts. The temporary and reactive nature of these pacts has, however, failed to place a strong constraint on wage-setters. In contrast to the Netherlands and Ireland, Italy (and to some extent Portugal) did not try to impose any legal constraints on their sheltered and exposed sectors. As a result both sectors increased wages to above productivity levels, thereby undermining their cost competitiveness.198 CONCLUSION INSTITUTIONAL INERTIA OR LEARNING OPPORTUNITY? This paper discussed the institutional issues associated with monetary, fiscal and labor instruments in the context of the EMU. We argued that Lisbon and Europe 2020 has provided an attempt to counterbalance EMU asymmetry through soft learning instruments such as the OMC. We further asserted that the OMC has been unable to counter the deregulatory pressures stemming from monetary and fiscal priorities. It was commonly agued199 that the weakness of the institutional progression of the EMU reflects a self-contradictory extension of the neoliberal Maastricht model. On the one hand, the EU imposed strict limits on national debt levels and on public deficits through the use of economic and monetary policy, and by subordinating social progress to economic success. On the other hand, it attempted to promote a learningbased approach to the evolution of social policy through the application of the OMC. While we acknowledge the validity of these criticisms and we recognize that fiscal and monetary policy did in fact exert pressure on national welfare states, the majority of these pessimistic predictions did not come to pass, at least until the onset of the crisis.200 It is only the GFC that brought the EMU’s vulnerabilities to the fore. Indeed, this article challenged the conventional wisdom that describes the crisis a fiscal profligacy issue. The economic turbulence the EU has experienced since 2007 did not in fact originate in the public purse. While the GFC has turned into a debt crisis the most significant contributing factor to this outcome was the asymmetric EMU governance that led to high dependence of the periphery on capital inflows and the rise of external – primarily private – debt. Contrary to popular belief, external debt in Greece and Portugal as well as in Spain and Ireland was mainly the result of private borrowing. There was only a tenuous link to public sector deficit, except in the case of Greece, which had a large deficit before 2007. Other countries, particularly Spain and 198

Hancké, supra note 156. W. Walters & J. H. Haahr, Governmentality and Political Studies, 4 EUR. POLIT. SCI. 288–300 (2005); K Mitchell, Neoliberal Governmentality in the European Union: Education, Training and Technologies of Citizenship, 24 ENVIRON. PLAN. SOC. SPACE 389–407 (2006); O. De Schutter & S. Deakin, Reflexive Governance and the Dilemmas of Social Regulation, 28 in SOCIAL RIGHTS AND MARKET FORCES: IS THE OPEN COORDINATION OF EMPLOYMENT AND SOCIAL POLICIES THE FUTURE OF SOCIAL EUROPE? (S. Deakin & O. De Schutter eds., 2005). 200 S. Dahan, The EU/IMF Financial Stabilisation Process in Latvia and Its Implications for Labour Law and Social Policy, 41 IND. LAW J. 305–327 (2012); S. Dahan, supra note 136. 199

42 Ireland, were, however, the main source of growth201 and examples of fiscal probity. The root cause of public deficit and debt was artificial and unsustainable growth in the private sector with leveraging and deleveraging, which required the States to assume the cost of deleveraging. The debt was initially private, incurred by banks and households, but it ultimately had to be assumed by governments after the financial shock of 2007. Furthermore, public sector deficit was worsened by the immediate effect of the crisis on fiscal receipts: tax receipts decreased while expenditure went up. Most significantly, this paper has raised serious questions about the legal origins of this fiscal crisis. We have argued that the legal asymmetry of the EMU has played a crucial role in the development of the euro crisis. Particularly, the uniform ECB monetary policy, the distorted SGP and the un-coordinated wage policy have amplified the development of imbalances. Firstly, while the ECB single-interest rates were too high in low-inflation States, they were too accommodating in high-inflation States, thus fuelling not only finance-led growth but also rapid wage inflation and competitiveness deterioration. Secondly, the numerical bias of the SGP prevented the Union from foreseeing or addressing the widening gap between the periphery and the core. Instead of fostering true convergence, the SGP facilitated the development of asset-price bubbles, excessive private debts and current account imbalances. Neither the ECB nor the Commission saw rising current-account deficits, here dependent on capital flows, as serious issues that required intervention. Furthermore, the SGP exacerbated the pro-cyclical dynamic of the EMU by rewarding States that showed surpluses and punishing those with deficits. The SGP was strictly obsessed with nominal figures and did not distinguish between deficits incurred between recessions and periods of growth. Finally, the absence of a common approach to wage setting aggravated this pro-cyclical dynamic by allowing competitiveness divergence. On the one hand, export-orientated States, where wage coordination remained strong, successfully constrained wage inflation. Consequently, they accumulated competitiveness gain and current account surpluses. On the other hand, finance-led States, where wage coordination was weak, were unable to mitigate wage inflation. Consequently, their export competitiveness deteriorated and they accumulated current account deficits. In sum, export-led strategies leading to current account surpluses were matched by current account deficits elsewhere, that is to say macroeconomic imbalances. While the phenomenon of current account imbalances is at least partially attributable to the lack of wage coordination and labor regulation, we do not argue for a fully-centralized EU collective bargaining process. As noted by Soskice and Iversen, this would be impractical.202 The crisis should, however, create a window of opportunity for realizing that wage policy – and, more broadly, social policy coordination – deserve more attention in future, as the lack of social policy coordination has contributed to the development of macroeconomic imbalances. However, as discussed elsewhere,203 neither the emergency- nor longer-term responses to the crisis treat social and employment policy as a priority. The disciplinary and austerity response only confirms the path-dependent limitations of 201

France or Germany were not the main source of growth. D. Soskice & T. Iversen, Multiple Wage-Bargaining Systems in the Single European Currency Area, 14 OXF. REV. ECON. POLICY 110–24 (1998). 203 Dahan, supra note 10. 202

43 the EMU asymmetric legal construction described in this article. We do not argue that the EMU governance system abandon institutional progression, especially since it has triggered a significant institutional progress at the onset of the crisis. The only problem is that it reinforces the domination of fiscal and monetary policy over social and labor policy: the ECB uses whichever instruments are available to assume a dominant position from which to address the crisis, and the Union reinforces the pre-crisis paradigm in order to ensure financial stability. This limited approach seems so far to have to have been the only possible path of action for the EU. While this solution has its merits, it may not be sufficient to address the long-term causes of the crisis. Indeed, the bond-buying program of the ECB will likely stabilize the euro in the short run, but the underlying problem is that there is very little growth in the periphery States and this will have knock-on effects on the EU economy as a whole, which relies significantly on intra-EU trade.204 The issue is not whether the ECB is doing something economically wrong. The problem is more one of an absence of growth and a ‘drifting’ of social policy resulting from austerity and the new economic governance framework, which in the long term might jeopardize the single currency as a whole. This limited form of learning is considered problematic mainly because it does not learn from past mistakes, nor does it address the real causes of the problem. This article claims that the EU might need to engage in a more genuine institutional evolution, one that challenges the paradigm under which the problem occurred in the first place. The new economic governance is only a partial step in that direction. While it tackles the issue of macroeconomic imbalances, it merely focuses on the symptoms of the crisis, namely deteriorating fiscal positions and high labor costs. In any case, the counterproductive impact of austerity might in the longer-term trigger a more genuine reflection on how the economic values underlying the EU constitution have laid the groundwork for the crisis and distorted its legal response. In sum, the crisis has created a window of opportunity in which to challenge the theoretical and ideological foundations of the EMU, and by the same token to design a more socially orientated approach. The role of lawyers is to investigate possible legal avenues towards achieving a triple-loop solution. One ambitious route would be to advocate a ‘great rebalancing’, a fully-fledged Social and Political Union that might require significant Treaty changes.205 A more modest solution would be to opt for a ‘reflexive rebalancing’ that does not have to take the form of hard law. In the long term, the EU might see the emergence of a third-order change in the form of a great rebalancing, as we have shown that the disciplinary response is not an adequate approach and that the crisis will deepen unless the markets see growth in the EU. In the short-term, however, a radical paradigmatic shift accompanied by a fully-fledged Union does not seem likely to occur, nor is it the only way to avoid disintegration. A more plausible solution would be to draw the correct lessons from the crisis, addressing the EMU asymmetric design while remembering the positive lessons of the OMC. Future research may thus investigate the potential a reflexive rebalancing solution that does not take the form of hard law. In particular, we argue that the EU requires 204 205

Germany exports to Asia but the majority of their export stays within the EU. Habermas, supra note 9.

44 neither new powers nor a new treaty to achieve institutional redirection. Treaty changes may help, but the EU already has several legal instruments at its disposal to counter internal devaluation pressures. The EU already has all the necessary legal means to implement a reflexive rebalancing solution through learning mechanisms.

45 Table 1 – Increase in Household Debt (in % GDP) 2000/2004 and 2000/2008 Export-led

2000–2004

2000–2008

Finance-led

2000–2004

2000–2008

Germany

–2.74

–11.34

Greece

18.26

Austria

7.05

7.21

Spain

22.01

Netherlands

24.35

29.1

Portugal

14.08

21.31

Italy

13.05

18.09

Ireland

35.07

61.72

32.53

Source: Eurostat206 Table 2 – Budget Deficit/GDP (%) 2006

2007

2008

2009

2010

2011

2012

France

-2.4

-2.7

-3.3

-7.6

-7.1

-5.7

-4.5

Germany

-1.7

0.2

-0.1

-3.2

-4.3

-1.2

-1.1

Greece

-6.0

-6.8

-9.9

-15.8

-10.8

-9.0

-7.0

Italy

-3.4

-1.6

-2.7

-5.4

-4.5

-3.6

-1.6

Portugal

-4.1

-3.2

-3.7

-10.2

-9.8

-5.9

-4.5

Spain

2.4

1.9

-4.5

-11.2

-9.3

-6.2

-4.4

-1.4

-0.7

-2.1

-6.4

-6.3

-4.0

-2.9

Euro area countries)

(15

Source: OECD

206

While household debt is falling in Germany and moderately increasing in Austria, it is dramatically increasing in the Southern periphery, with all countries well above the Euro (12) area average. In the Netherlands household debt increases rapidly as well, though not as fast as in Ireland and Spain.

46 Table 3 - Public Debt/GDP (%) and Change in Debt in 2010-2007 (%-points) 2006

2007

2008

2009

2010

2011

2012

France

71.2

73.0

79.3

90.8

95.2

98.6

102.4

Germany

69.8

65.6

69.7

77.4

87.1

86.9

87.3

Greece

116.9 115.0 118.1 133.5 149.1 165.1 181.2

Ireland

29.2

Italy

116.9 112.1 114.7 127.1 126.1 127.7 128.1

Portugal

77.6

75.4

80.7

93.3

103.6 111.9 121.9

Spain

46.2

42.3

47.7

62.9

67.1

74.1

77.2

Euro area (15 74.7 countries)

71.8

77.0

87.6

92.9

95.6

97.9

Source: OECD

28.7

49.6

71.1

98.5

112.6 118.8

47 Table 4 – Collective Bargaining Coverage Collective bargaining coverage rate A.

Country

Year

Proportion of wage and salaried earners

B.

C.

Proportion of total employment

Reported Proportion

Europe Belgium

2007

*96.0

France

2004

*97.7

Germany

2006

35.8

Italy

2004

*98.2

*96.0

Latvia

2006

34.7

39.9

Luxembourg

2007

49.8

Norway

2004

75.1

Poland

2008

*14.4

11

Portugal

2007

38.7

29.2

Spain

2006

68.6

Switzerland

2008

46.9

United Kingdom

2007

35.1

46.7

48

*53.9 74

70 36.9

32 34.6

*Denotes private sector coverage only. #Denotes

public sector coverage only.

Source: Trade union density and collective bargaining coverage: International Statistical Inquiry 2008-09, ILO 2010

48 Table 5 – Household Disposable Income as % of Nominal GDP 2004 2005 2006 2007 2008 2009 2010 2011 Germany Euro thirteen

area

OECD - Total

of

68.7

71.1

69.2

65.3

68.8

77.4

82.0

85.5

75.8

76.8

74.4

70.9

73.2

81.8

88.3

93.2

74.3

75.9

74.6

73.1

78.4

90.0

97.4

103.5

Source: OECD

Table 6 – Wage-Coordination Institutions and Policies

49 Figure 1 – Consumer Price Inflation

Source: OECD Figure

2

Source: OECD



Interest

Rates

on

Ten-Year

Government

Bonds

50 Figure 3 – Real Interest Rates

Source: OECD. Own calculation Figure 4 – Gross Domestic Product (GDP): GDP, Volume – Annual Growth Rates in Percentage

Source: OECD

51 Figure 5 – Current Account % of GDP

Source: OECD Figure 6 – Government Budget Deficit or Surplus as % of GDP

Source: Ameco

Figure 7 – Strictness of Employment Protection (overall)

52

Source: OECD

Figure 8 - Union Density in the Eurozone

Source: OECD and J.Visser, ICTWSS database (Institutional Characteristics of Trade Unions, Wage Setting, State Intervention and Social Pacts, 1960-2010), version 3.0 (http://www.uva-aias.net/)

53 Figure 9 – Wage Restraint under ERM and EMU Wage restraint is the change in nominal wage growth minus the change labour productivity. A negative outcome indicates wage restraint; a positive outcome indicates wage excess.

Source: Nominal wage growth data from AMECO and labour productivity growth data from OECD. Calculation by Hancke and Johnston 2009. Figure 10 – Unemployment Rates (Age 15-64)

Source: OECD

54 Figure 11 – Nominal Compensation per Employee(Annual Growth Rate) 20

15

10

Germany Greece

5

Ireland Portugal Spain

0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 -5

-10

Source OECD Figure 12 – Strictness of Employment Protection (Temporary Employment)

Source: OECD

Figure 13 – Unit Labour Cost (Total Economy, Annual Growth Rate)

Source: OECD Ratio of compensation to annual growth rate per person employed